Live updates to machine learning models

Abstract

To enable resources, such as machine learning models, to be updated for an application without any significant downtime for that application.SOLUTION: For an application hosted at a network edge, the application can be deployed in a container, and one or more model versions can be stored in local storage at the edge and can be mounted on the container as necessary. When a different model version is to be used, a configuration change or new context can be used to trigger the application to automatically change to the different model version. This update can be performed seamlessly, without any loss of data.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] This application claims priority to U.S. Provisional Patent Application No. 62 / 923,591, entitled "A Method of Performing Live Updates of Machine Learning Models While in Use," filed October 20, 2019, which is hereby incorporated by reference in its entirety for all purposes. [Background technology]

[0002] Techniques such as machine learning are increasingly being used for a variety of tasks across a wide range of industries. For applications that utilize deep learning models, these models are typically deployed along with the application, such as in an application container image. To achieve the highest level of accuracy for machine learning-based inference, it may be desirable to frequently update these models. Summary of the Invention [Problem to be solved by the invention]

[0003] However, deploying such new or updated models using traditional techniques requires deploying new containers, which must stop old versions of the applications from running before the new versions can run instead, which can cause significant downtime for those applications or can cause those applications to become unavailable.

[0004] Various embodiments according to the present disclosure will now be described with reference to the drawings. [Brief description of the drawings]

[0005] [Figure 1] FIG. 1 illustrates an example architecture for supporting edge applications, according to at least one embodiment. [Diagram 2]FIG. 1 illustrates a model management system, according to at least one embodiment. [Diagram 3] FIG. 1 illustrates an example process for updating a machine learning model for an application, according to at least one embodiment. [Figure 4] FIG. 1 illustrates an example process for launching an edge application with a machine learning model, according to at least one embodiment. [Diagram 5] FIG. 1 illustrates an example process for updating a machine learning model for an edge application, according to at least one embodiment. [Figure 6] FIG. 1 illustrates an example process for changing the version of a model utilized by an edge application, according to at least one embodiment. [Figure 7A] FIG. 1 illustrates inference and / or training logic, according to at least one embodiment. [Figure 7B] FIG. 1 illustrates inference and / or training logic, according to at least one embodiment. [Figure 8] FIG. 1 illustrates an example data center system in accordance with at least one embodiment. [Figure 9] FIG. 1 illustrates a computer system according to at least one embodiment. [Figure 10] FIG. 1 illustrates a computer system according to at least one embodiment. [Figure 11] A diagram illustrating at least a portion of a graphics processor in accordance with one or more embodiments. [Figure 12] A diagram illustrating at least a portion of a graphics processor in accordance with one or more embodiments. [Figure 13] FIG. 1 is an example data flow diagram for an advanced computing pipeline, according to at least one embodiment. [Figure 14] A system diagram of an example system for training, adapting, instantiating, and deploying machine learning models in an advanced computing pipeline, according to at least one embodiment. [Figure 15A] FIG. 1 is a data flow diagram of a process for training a machine learning model, according to at least one embodiment. [Figure 15B] FIG. 1 illustrates a client-server architecture for extending annotation tools with pre-trained annotation models, according to at least one embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0006] Techniques such as machine learning are increasingly becoming relied upon for a variety of different tasks. In many applications, due in part to the resources required to perform machine learning, it may be desirable to have the machine learning hosted on a remote server or other such system rather than on the client device. While various systems may host machine learning and other such applications or functions, for example, from within a provider's network or environment, in various situations it may be desirable to host the machine learning at the edge of the network. Running inference applications at edge locations can have various advantages for customers, such as reduced latency, because edge locations can be near connected media sources and sensors, which can be important for latency-sensitive applications or servers that may be accessed from a variety of different geographic locations.

[0007] In certain use cases, this machine learning or artificial intelligence (AI) based functionality may need to operate continuously, such as while processing data from a video stream to generate inference outputs. In high demand applications, this use case may include the application being able to run continuously and without any gaps or downtime. Examples of such deployments can be found in applications that perform image or sensor data processing, object detection, identification, and / or classification, area monitoring or surveying, and world space perception, among others. Such applications are then used in a wide range of industries, including but not limited to retail stores, warehouses, airports, parking lots, and highway surveillance. Tasks such as inference, as well as downstream analytics computed from the output of this inference, may be optimally performed using a particular type of hardware, such as may include a server with one or more graphics processing units (GPUs), where the application may be containerized. Often, the deployment of the servers and applications is managed by a container orchestration platform, such as the open source Kubernetes platform. In at least some embodiments, an application container may include or be composed of an entire runtime environment for that application, including various libraries, binaries, dependencies, and configuration files for that application.

[0008] Various systems can perform inference on video or media streams by running video frames through machine learning or neural network models that achieve a certain accuracy and throughput performance. However, more accurate and faster models are constantly being developed to improve the performance of these inference applications. In traditional approaches, the models are typically included (i.e., pre-packaged) in the container image of the underlying application that uses the inference output. However, this means that a new container image, which represents a new version of the application, must be built and distributed to the compute nodes at edge locations in order to update to a newer and better model. Introducing a new version of the application means that the currently running container on the node must be stopped and a new container must be started, which results in downtime for inference and analysis. This downtime can be roughly 60 seconds or more in some instance situations, depending on factors such as the complexity and structure of the application and the capabilities of the compute nodes. Thus, operators of these inference nodes face the possibility of experiencing frequent outages, gaps in inference or analysis, or the use of outdated or suboptimal neural network models.

[0009] Thus, the techniques according to various embodiments allow models to be updated with little or no downtime, without data loss, and without the need to update or restart applications. Various objects, elements, algorithms, processes, or code may be updated using features well within the scope of various embodiments. In at least one embodiment, models are not built into the container. Instead, these models can be mounted into the container from storage of the system on which the container is running. In various embodiments, a service component can be utilized that can pull updated model versions into the local storage of the system or edge node. In at least one embodiment, an edge manager mechanism is then used to signal the running inference container application to switch to a different model version. Such techniques allow models to be updated using a live update process for models in running application containers without the need to break the container or drop any input data to be processed. Such models can take various forms and, at least some embodiments, may have encryption or security mechanisms applied that can be managed by the techniques presented herein.

[0010] FIG. 1 illustrates an example architecture that may be utilized to provide such functionality, according to at least one embodiment. In implementations in which inference is utilized at the edge for one or more video streams, these streams may include video and other media content that may be presented from a media source 160, which may include, but is not limited to, a digital camera, a video camera, a medical scanner, any of a computing device, or other such content source. The media provided from the media source 160 may be presented using a client device 120, which may include a desktop computer, a notebook computer, a set-top box, a streaming device, a gaming console, a smart phone, a tablet computer, a virtual reality (VR) headset, an augmented reality (AR) goggles, a wearable computer, or a smart television, etc. In at least one embodiment, the content transmitted by the media source 160 may include video frames. In at least one embodiment, the content transmitted by the media source 160 may include medical scan data, such as DICOM images. In at least one embodiment, the media source is connected through a network to an edge server 120. In at least one embodiment, an inference application 124 executing on the edge server 120 can initiate a session associated with the client device 102, such as by using a session manager and user data from a user account, and can send content to the client device 102 using one or more video streams for that session, which may be managed, such as by using an appropriate stream manager 122.In at least one embodiment, a client device 102 receiving this content can provide this content for presentation via one or more components of the client device 102 or one or more components in communication with the client device 102, which may include a display 106 for displaying image or video content (which may include animation and gaming content). In at least one embodiment, a transmission mechanism other than streaming may also be used to transfer at least a portion of this content from the edge server 120 or another such source. The analysis application 104 executing on the client device 102 can perform various analyses on the content received by the client device 104, as described herein.

[0011] In at least one embodiment, the content to be streamed may be stored in storage 132 of the edge server and may be generated from a live video stream using one or more video cameras or one or more other imaging devices and using one or more imaging modalities. In one or more other embodiments, the content is received from one or more media sources 160. In other embodiments, the content may instead be generated at cloud server 140, edge server 120, or another suitable location, such as by using a content generator application 136 for generating gaming, animation, or other such content. This content may also be received over one or more networks 110, in at least some embodiments. In at least one embodiment, the content may be received by an inference application 124, which may perform inference on the content before transmitting the content (e.g., a video stream) to client device 102. Inference on live video streams may be used for a variety of purposes, such as security, medical imaging, traffic monitoring, or inventory control, among other such options. Inference at the edge can be performed on a variety of other types of data and inputs well within the scope of various embodiments. As mentioned above, there may be one or more deep learning (DL) models 128, or neural network models, available to the inference application 128, which may be stored at the edge server 120, at least initially, in storage 132 outside the application container 126, where they may be mounted to the application container 126 using the inference application controller 130. In at least some embodiments, the model to be used may be determined by reference to a manifest 134 that identifies the current state or information of the inference application, where the manifest 134 may be updated by a user or other source.

[0012] In at least one embodiment, additional or new models may be received by each appropriate edge server 120 from a central source, such as a provider cloud server 140 that is behind a firewall or otherwise at least partially separated from the edge servers 120 and the client devices 102. This cloud server 140, or other such system or service, may store different versions of one or more models in a model repository 144, for example, which may push such models to the edge server 120 or enable the edge server 120 to obtain a particular model, such as by enabling the edge server to request a particular model that may be identified in the current manifest 134. The edge server may then reference the manifest 134 to ensure that the appropriate model, and the appropriate version of that model, is used for inference on the content to be sent to the client device 102.

[0013] FIG. 2 illustrates a component 200 that can be utilized in an architecture such as that of FIG. 1 to update and manage models, according to various embodiments. As discussed above, in such embodiments, deep learning models can be separated from applications, but rather can be mounted into application container 220 at runtime from host system storage. The host system stores models in local storage 210 with directories 212A-212C for each model, each of which can include one or more subdirectories for different versions of that model. These model directories can be mounted into application container 220 at runtime, either individually or as part of an overall model directory, such that the directories become visible to inference application 202, which can begin execution with an initial version of each model, which can be identified by manifest file 216. In at least one embodiment, a model fetcher software component can fetch new versions of any of these models from model store 206 or other such location, and these versions can be made available over at least one network as they become available. A model fetcher, which can mount the model directory via a read / write mount, can then store these versions in a new subdirectory of the associated model directory 212. An application management system, such as edge manager 208, can then signal the inference application 202 to switch to the new version of the model. An administrator user interface 222, in communication with edge manager 208, can allow an administrator or other authorized entity to specify or update version-related information or other such information. The application 202 may create a new context that is loaded with the new model version.In at least some embodiments, the context may include a software inference engine. When the new context is ready to operate, the inference application 202 may switch inference operations from the old context to the new context, and the old context may then be deleted. Inference can then be performed using the new model version without interruption, and downtime during this model update is reduced to zero or near zero.

[0014] In example implementations according to various embodiments, the deployed edge computing application may include a deep learning inference platform to load their models, such as NVIDIA's TensorRT (TRT) or Triton Inference Server (Triton) platforms. In such embodiments, the models may be stored in local storage with a directory hierarchy for different versions, or in locally mounted storage. Model conversion to fit a particular GPU and format (e.g., T4 engine files) may be performed offline, whereby the exact format required by the GPU can be written to a directory already mounted in the running container. The edge computing application can monitor the mounted configuration map and react to changes using methods such as inotify (Linux inode notify). The edge computing application can then be modified to switch to the new model version without an application restart. TensorRT (TRT) is a runtime optimized for AI inference on GPUs. In one or more embodiments, the edge computing application can use TRT to load the neural network model. Triton provides a cloud inference solution optimized for NVIDIA GPUs, which may be embedded in applications. In one or more embodiments, edge computing applications may use Triton to load neural network models. The EGX stack / EGX node stack is a software stack that may run on GPU-enabled servers (e.g., NVIDIA's EGX system) running edge use cases such as AI inference. The EGX Management Server (EMS) may be used as a service to manage the EGX system as an AI appliance on behalf of a customer or operator.The Edge Manager (EM) acts as an interface to the EMS and can be used as an application management system in multiple forms: API, WebUI, and CLI.

[0015] Such an approach can provide various benefits for different use cases according to various embodiments. For example, such an approach can provide a separation of application and model, since an application developer can provide an application without a model or with a default model, and can attach the application to the model or an updated model on the target system. This allows the development and release of the application and the model to be decoupled, reducing the size of the application container image and the frequency of releases or updates. A developer, data scientist, or other such source can then release a new version of the model and make it available to the edge application without requiring a change or release of the application. In some cases, an edge application administrator may make newly released versions of the model available from a location such as cloud storage to facilitate deployment to multiple edge locations. An edge application administrator can also generate or provide newly released versions of the model to be automatically fetched to edge locations where applications that use these models run. An edge application administrator can choose when to perform model updates at each edge location, and can update the model to the new version without downtime in running applications. Edge application administrators can also roll back edge applications to a previous or other version of a model with no or near-no downtime.

[0016] For applications that wish to participate in live model updates, models may be removed from the container image and a volume mount to a specific directory (e.g., / models) may be documented at container startup. An alternative approach is to ship with a default version of each model in a separate (e.g., default) directory, which can be used if the model directory is not created as a mounted volume at startup. Models separated from applications may be released in semantic versioning ("semver") versions, allowing edge application administrators to distinguish between release ordering and compatibility breaks. In one or more embodiments, models may follow a naming and versioning scheme defined in a model registry, such as NVIDIA's NVIDIA GPU Cloud (NGC) Model Registry. To be made available for pulling to edge locations, models may be hosted in a software hub, such as NVIDIA GPU Cloud (NGC). In NGC, models may be public and guest accessible, or may be in a private area that requires a customer account and credentials (e.g., API key) to access.

[0017] In at least some embodiments, a model fetcher 204 can be used to fetch models from a remote model store 206 or software hub, such as NGC, to an edge computing system (e.g., EGX) at an edge location as shown in FIG. 2. According to one or more embodiments, the model fetcher 204 can read a model manifest file 216, which may list model repositories to monitor, such as in NGC, and may list specific versions to download from the model store 206 when they become available, and the version specifications may be wildcarded to insert "wildcards" or other characters (e.g., "*") that can then match future characters or sequences of future characters for one or more versions. The model manifest file 216 can also identify a destination directory 212, such as on an EGX node, where the models will be stored and mounted in the application container. In some embodiments, the model fetcher 204 may periodically scan the model manifest file 216, for example by checking the NGC registry for matching model versions. Any model versions that are not in the destination directory may be fetched and stored. The model fetcher 204 may also periodically call the EMS API to provide a current list of available models and versions.

[0018] In at least one embodiment, the edge manager 208 can specify the model and model version to be used by the application as part of the deployment configuration. If the edge manager 208 updates a deployment and the only update includes one or more model version differences, the application can identify the update and switch to the new model version without terminating. In at least one embodiment, a deployment update can be performed by updating a configuration map (e.g., configmap) mounted to the associated application container. The application can detect that the configmap has been updated, such as by using inotify. When the application detects that a change in model version is being requested, the application stops using the current model version and begins using the new model version, which in some embodiments may always point to the latest version. In at least one embodiment, this can be done by creating a new software context for the application, into which the new model is loaded. Once the new software context is ready, the inference stream is switched to the new context and the old context is deleted. Similar to deployment updates, deployments of model updates can be rolled back, such as by using Helm functions. In one or more embodiments, the application may again recognize that the only change in its configuration is the model version, and may repeat the model version switching steps described above to switch to the previous version. In at least one embodiment, other applications that are more peripheral than the model may also be dynamically switched using such techniques, which may relate to graphics resources, etc.

[0019] In at least some embodiments, the process of updating the neural network model for an edge inference application may require that an image of the edge inference application (e.g., NVIDIA's DeepStream) be deployed with one or more neural network models to perform the inferences contained in the container image. According to such embodiments, the edge inference application may use a deep learning inference platform such as TensorRT to manage the models. The local storage 210 containing the one or more models may be mounted in a read-only manner to the edge application container 220. In one or more embodiments, the edge application 202 (e.g., DeepStream) may use an edge manager 208 such as TensorRT to manage the models. In one or more embodiments, the local storage 210 of the edge node (e.g., EGX node) may be manually provisioned with the models. In at least one embodiment, an analysis server 214 or client may communicate with the container 220 to perform various analyses described herein.

[0020] A local storage 210, such as a local storage volume, may be populated with versioned models in various model directories 212. In one or more embodiments, the application 202 may use an edge manager 208 to manage models as well as execute models stored in the mounted local volumes. According to such embodiments, the local storage of the edge node may include separate directories for each model, which may have separate subdirectories for each model version. A model fetcher 204 component may move models to the edge node from a model store 206, such as a cloud repository or other such source. In at least one embodiment, a stack component of the edge node may fetch models from the model store 206 to the edge node and write them to the appropriate model directory 212 via a read / write mount. Such an architecture may also work for applications that use other approaches to manage models, such as Triton. For example, another edge inference application, such as Clara by NVIDIA, may use Triton to manage models.

[0021] In at least one embodiment, the model fetcher 204 can perform tasks such as managing volume mounts of models for each application. The model fetcher can fetch models from the model store 206 according to a manifest and at least one policy, which can indicate how often to check for updates (e.g., daily or hourly). The model fetcher 204 can convert the models to an appropriate format (e.g., Triton) as needed and write each model to an appropriate versioned directory or subdirectory for that model. The model fetcher 204 can also maintain an inventory of all models and versions stored at a given edge node or in a given storage volume. In an embodiment where multiple servers are at an edge location, each of these servers can pull the model version, such as by using a given URL, or the model can be pulled or pushed once to a mounted volume that can point to all three servers so that they have the model version locally.

[0022] FIG. 3 illustrates an example process 300 for updating a model version used by an application that may be utilized by various embodiments. It should be understood that there may be additional, alternative, or fewer steps performed in a similar order or at least in an alternative order or in parallel with respect to the above and other processes discussed herein within the scope of various embodiments unless otherwise noted. In this example, an application is executed 302 using a first version of a machine learning model. The application may be an inference application that executes at a data center, at a network edge, or at another suitable location. While the application is executing, a second version of the machine learning model may be received 304 or otherwise obtained from a source such as a model provider or a model repository. New configuration data may be received 306, generated, or otherwise obtained that specifies that the application should use the second version. In response, the application may be automatically switched 308 to use the second version, with little or no downtime for the inference application. In this example, the first version of the model may then be removed from at least the container or other environment associated with the inference application.

[0023] FIG. 4 illustrates an example process 400 for deploying a model for an inference application that may be utilized by at least one embodiment. In this example, an application container image may be deployed 402 to an edge server, which may be one of multiple edge servers that will host the inference application associated with the container. The edge server's manifest may be updated 404 to specify the machine learning model to be used for the inference application when running on the edge server. Information about the model may be found 406 in the manifest, and the corresponding model (and version, etc.) may be retrieved from a model store or other such source and stored in a local directory, such as a model-specific directory or subdirectory of a local storage volume. At the appropriate time, the application container may be launched 408 at the edge server. As part of the launch process in this example, the model may be mounted 410 from the local directory to the application container for use in the inference application. The application may then identify and switch to using the model 412 as appropriate, and inference is performed using the particular model. In an embodiment where a container may be deployed with a default model, an inference application can switch to a specific model with little to no downtime.

[0024] FIG. 5 illustrates an example process 500 for updating a model to be used in an inference application, such as that described in process 400 of FIG. 4. In this example, an updated (or other different) version of a model to be used in an inference application at one or more edge servers, or other such locations, is generated 502. An application manifest may be updated 504 at each edge server to specify the new version of the model (or new model, if applicable) to be used in the inference application. At each edge server, the manifest may be analyzed 506 and information about the new version may be detected. The new version of the model may then be retrieved 508 from a remote model store or other such location, and the version may be stored in a model directory or subdirectory of the edge server's or node's local storage. In at least some embodiments, the model may be retrieved by a model fetcher at the edge server. The new version of the model may be mounted 510 to the application container along with new configuration information. New or updated configuration information or application context that corresponds to or points to the new version of the model may be created 512 or otherwise obtained. The inference application, while running on the edge server, can detect the new configuration information and automatically switch to the new version of the model with little impact to the availability of the inference application 514. This can include the application detecting the new configuration information or context, determining that the new context is securely loaded and prepared with appropriate buffers, and then switching to the new model version without dropping frames or other instances of input. In at least some embodiments, the previous version of the model can be removed from the application container. The inference application can then be allowed to run using the new version of the model 516.If it is determined 518 that a new model exists, the application manifest may be updated to specify the new version and the steps of the process of updating the model may be repeated.

[0025] FIG. 6 illustrates an example process 600 for switching to a previous version of a model that may be utilized by various embodiments. This process 600 may be utilized after updating to a new version of a model utilizing a process such as that described with respect to FIG. 5. In this example, it is determined 604 to use the previous version of the model, such as when the current version does not perform as expected, or when it is determined that a different version should be utilized for a particular edge server or other deployment. This may also be the case when different versions are utilized at different times, for different workloads, or for other such reasons. In this example, it is verified 604 that the previous version of the model exists in a manifest. The manifest may be analyzed 606 to obtain information about the previous version. In some examples, a check may be performed to verify that the previous version is still in local storage, or if not, whether the previous version is available from a model store or other such source. Updated configuration information (e.g., application context) corresponding to the previous version of the model may be created 608. The inference application may detect the updated configuration information and automatically switch 610 to the previous version of the model. In at least one embodiment, the version of the model that is no longer being used may then be removed from the application container and the inference application may then be allowed to run 612 using this version of the model.

[0026] As discussed above, such processes can be used to revert to a previous version if a potential problem is detected with the new version, or can be used to switch between versions of a model at various times, such as when a particular version is determined to be preferable for the current model being used for inference. In at least some embodiments, each version of one or more models to be used for an edge application can be pushed to each edge server hosting that application, and these applications can then switch versions using the locally stored version. While the decision to switch versions can be made remotely, the actual switch can be performed locally at the edge without any downtime, sideband communication, or additional steps. As discussed above, this can also result in different applications using different versions of a model at different times or under different conditions. There can be one or more policies in place indicating which model should be used at what time or under what conditions, and these policies can be used to enable local model switching, in at least some embodiments. One example non-limiting policy can be to always use the latest model or version. In some instances, context or configuration information can be stored in local memory for multiple model versions, and this stored configuration information can be used to dynamically switch context to switch the application between different versions.

[0027] As mentioned above, different container technologies such as Kubernetes or Docker may be used in different embodiments. For a Kubernetes-based approach, a Kubernetes configuration map mechanism may be used to update a configuration file accessible to an application running at the edge. This configuration file change may act as a signaling mechanism that something has changed, allowing the application to determine the model version that should be used and automatically switch to that version. In this example, the configuration file change may be triggered by a configuration mapping change. Kubernetes automatically updates the application configuration file in response to such a configuration mapping change. In at least one embodiment, one or more APIs may be exposed, such as by a remote administrator console, that allow this configuration mapping to be updated. Such a process may also be used for resources other than models, such as to update textures, maps, algorithms, modules, or audio files for a game hosted at the network edge to different versions without modifying the application code or causing downtime in its execution.

[0028] Logic of inference and training 7A illustrates inference and / or training logic 715 that may be used to perform inference and / or training operations for one or more embodiments. More details regarding inference and / or training logic 715 are provided below in conjunction with FIGs. 7A and / or 7B.

[0029] In at least one embodiment, the inference and / or training logic 715 may include, without limitation, code and / or data storage 701 for storing forward and / or output weights, and / or input / output data, and / or other parameters for configuring neurons or layers of a neural network that are trained and / or used to infer in one or more aspects of the embodiment. In at least one embodiment, the training logic 715 may include or be coupled to code and / or data storage 701 for storing graph code or other software for controlling the timing and / or ordering of the weights and / or other parameter information to configure logic including integer and / or floating point units (collectively arithmetic logic units (ALUs)). In at least one embodiment, code such as graph code loads the weights or other parameter information into a processor ALU based on the architecture of the neural network to which the code corresponds. In at least one embodiment, code and / or data storage 701 stores weight parameters and / or input / output data for each layer of a neural network trained or used in conjunction with one or more embodiments during forward propagation of the input / output data and / or weight parameters during training and / or inference using aspects of one or more embodiments. In at least one embodiment, any portion of code and / or data storage 701 may be included with other on-chip or off-chip data storage, including L1, L2, or L3 caches of a processor, or system memory.

[0030] In at least one embodiment, any portion of the code and / or data storage 701 may be internal or external to one or more processors or other hardware logic devices or circuits. In at least one embodiment, the code and / or code and / or data storage 701 may be cache memory, dynamic randomly addressable memory (“DRAM”), static randomly addressable memory (“SRAM”), non-volatile memory (e.g., flash memory), or other storage. In at least one embodiment, the choice of whether the code and / or code and / or data storage 701 is internal or external to the processor, for example, or comprised of DRAM, SRAM, flash, or some other type of storage, may depend on the storage available on-chip versus off-chip, the latency requirements of the training and / or inference functions being performed, the batch size of data used in the inference and / or training of the neural network, or any combination of these factors.

[0031] In at least one embodiment, the inference and / or training logic 715 may include, without limitation, code and / or data storage 705 for storing backpropagation and / or output weights and / or input / output data corresponding to neurons or layers of a neural network trained and / or used to infer with one or more aspects of the embodiment. In at least one embodiment, the code and / or data storage 705 stores weight parameters and / or input / output data for each layer of a neural network trained or used in conjunction with one or more embodiments while backpropagating input / output data and / or weight parameters during training and / or inference using aspects of one or more embodiments. In at least one embodiment, the training logic 715 may include or be coupled to code and / or data storage 705 for storing graph code or other software for controlling the timing and / or sequence, where the weights and / or other parameter information are loaded into logic including integer and / or floating point units (collectively arithmetic logic units (ALUs). In at least one embodiment, code such as the graph code loads the weights or other parameter information into the processor ALUs based on the architecture of the neural network to which the code corresponds. In at least one embodiment, the training logic 715 may include code and / or data storage 705 for storing the graph code or other software for controlling the timing and / or sequence, where the weights and / or other parameter information are loaded into the processor ALUs. In an embodiment, any portion of code and / or data storage 705 may be included with other on-chip or off-chip data storage, including L1, L2, or L3 cache of a processor, or system memory. In at least one embodiment, any portion of code and / or data storage 705 may be internal or external to one or more processors, or other hardware logic devices or circuits. In at least one embodiment, code and / or data storage 705 may be cache memory, DRAM, SRAM, non-volatile memory (e.g., flash memory), or other storage.In at least one embodiment, the choice of whether the code and / or data storage 705 is internal or external to the processor, for example, or comprised of DRAM, SRAM, flash, or some other type of storage, may depend on the storage available on-chip versus off-chip, the latency requirements of the training and / or inference functions being performed, the batch size of data used in inference and / or training of the neural network, or any combination of these factors.

[0032] In at least one embodiment, code and / or data storage 701 and code and / or data storage 705 may be separate storage structures. In at least one embodiment, code and / or data storage 701 and code and / or data storage 705 may be the same storage structure. In at least one embodiment, code and / or data storage 701 and code and / or data storage 705 may be partially the same storage structure and partially separate storage structures. In at least one embodiment, any portion of code and / or data storage 701 and code and / or data storage 705 may be included with other on-chip or off-chip data storage, including a processor's L1, L2, or L3 cache, or system memory.

[0033] In at least one embodiment, the inference and / or training logic 715 may include one or more arithmetic logic units ("ALUs") 710, including without limitation integer and / or floating point units, for performing logical and / or arithmetic operations based at least in part on or indicated by the training and / or inference code (e.g., graph code), the results of which may generate activations (e.g., output values ​​from layers or neurons in a neural network) that are stored in activation storage 720 and that are functions of input / output and / or weight parameter data stored in code and / or data storage 701 and / or code and / or data storage 705. In at least one embodiment, the activations stored in activation storage 720 are generated according to linear algebra and / or matrix based calculations performed by ALU 710 in response to executing instructions or other code, where weight values ​​stored in code and / or data storage 705 and / or code and / or data storage 701 are used as operands along with other values, such as bias values, gradient information, momentum values, or other parameters or hyperparameters, any or all of which may be stored in code and / or data storage 705, or code and / or data storage 701, or in another storage, on-chip or off-chip.

[0034] In at least one embodiment, ALU 710 is included within one or more processors or other hardware logic devices or circuits, while in other embodiments ALU 710 may be external to the processors or other hardware logic devices or circuits that use them (e.g., a coprocessor). In at least one embodiment, ALU 710 may be included within an execution unit of a processor, or may be otherwise included within an ALU bank accessible by execution units of a processor, either within the same processor or distributed among different processors of different types (e.g., central processing units, graphics processing units, fixed function units, etc.). In at least one embodiment, code and / or data storage 701, code and / or data storage 705, and activation storage 720 may be in the same processor or other hardware logic devices or circuits, while in other embodiments they may be in different processors or other hardware logic devices or circuits, or some combination of the same processor or other hardware logic devices or circuits and different processors or other hardware logic devices or circuits. In at least one embodiment, any portion of activation storage 720 may be included with other on-chip or off-chip data storage, including the processor's L1, L2, or L3 cache, or system memory. Additionally, inference and / or training code may be stored with other code accessible to the processor or other hardware logic or circuitry, and may be fetched and / or processed using the processor's fetch, decode, schedule, execute, retire, and / or other logic circuitry.

[0035] In at least one embodiment, activation storage 720 may be cache memory, DRAM, SRAM, non-volatile memory (e.g., flash memory), or other storage. In at least one embodiment, activation storage 720 may be fully or partially internal to or external to one or more processors or other logic circuitry. In at least one embodiment, the choice of whether activation storage 720 is internal or external to a processor, for example, or comprised of DRAM, SRAM, flash, or some other type of storage, may depend on available on-chip versus off-chip storage, latency requirements of the training and / or inference functions being performed, batch sizes of data used in inference and / or training of a neural network, or any combination of these factors. In at least one embodiment, the inference and / or training logic 715 shown in Figure 7A may be used in conjunction with an application-specific integrated circuit ("ASIC"), such as a Tensorflow® processing unit from Google, an inference processing unit (IPU) from Graphcore™, or a Nervana® (e.g., "Lake Crest") processor from Intel Corp. In at least one embodiment, the inference and / or training logic 715 shown in Figure 7A may be used in conjunction with other hardware, such as central processing unit ("CPU") hardware, graphics processing unit ("GPU") hardware, or a field programmable gate array ("FPGA").

[0036] FIG. 7B illustrates inference and / or training logic 715, according to at least one or more embodiments. In at least one embodiment, the inference and / or training logic 715 may include, without limitation, hardware logic in which computational resources are dedicated to or otherwise used only in conjunction with weight values ​​or other information corresponding to one or more layers of neurons in a neural network. In at least one embodiment, the inference and / or training logic 715 illustrated in FIG. 7B may be used in conjunction with an application specific integrated circuit (ASIC), such as a Tensorflow® processing unit from Google, an inference processing unit (IPU) from Graphcore™, or a Nervana® (e.g., “Lake Crest”) processor from Intel Corporation. In at least one embodiment, the inference and / or training logic 715 illustrated in FIG. 7B may be used in conjunction with other hardware, such as central processing unit (CPU) hardware, graphics processing unit ("GPU") hardware, or field programmable gate arrays (FPGAs). In at least one embodiment, the inference and / or training logic 715 may include, without limitation, code and / or data storage 701 and code and / or data storage 705, which may be used to store code (e.g., graph code), weight values, and / or other information including bias values, gradient information, momentum values, and / or other parameter or hyperparameter information. In at least one embodiment shown in FIG. 7B, each of the code and / or data storage 701 and code and / or data storage 705 is associated with dedicated computational resources, such as computational hardware 702 and computational hardware 706, respectively. In at least one embodiment, each of the computational hardware 702 and computational hardware 706 includes one or more ALUs that perform mathematical functions, such as linear algebraic functions, only on the information stored in the code and / or data storage 701 and code and / or data storage 705, respectively, with the results stored in activation storage 720.

[0037] In at least one embodiment, each of the code and / or data storage 701 and 705 and corresponding computational hardware 702 and 706 correspond to a different layer of a neural network, such that activations resulting from one “storage / computation pair 701 / 702” of the code and / or data storage 701 and computational hardware 702 are provided as inputs to a “storage / computation pair 705 / 706” of the code and / or data storage 705 and computational hardware 706 to reflect the conceptual organization of the neural network. In at least one embodiment, the storage / computation pairs 701 / 702 and 705 / 706 may correspond to more than one layer of the neural network. In at least one embodiment, additional storage / computation pairs (not shown) may be included in the inference and / or training logic 715 after or in parallel with the storage / computation pairs 701 / 702 and 705 / 706.

[0038] Data Center 8 illustrates an example data center 800 in which at least one embodiment may be used. In at least one embodiment, the data center 800 includes a data center infrastructure layer 810, a framework layer 820, a software layer 830, and an application layer 840.

[0039] 8, in at least one embodiment, data center infrastructure layer 810 may include a resource orchestrator 812, grouped computing resources 814, and node computing resources ("node CRs") 816(1)-816(N), where "N" represents any positive integer. In at least one embodiment, node CRs 816(1)-816(N) may include, but are not limited to, any number of central processing units ("CPUs") or other processors (including accelerators, field programmable gate arrays (FPGAs), graphics processors, etc.), memory devices (e.g., dynamic read-only memory), storage devices (e.g., solid state drives or disk drives), network input / output ("NW I / O") devices, network switches, virtual machines ("VMs"), power modules, and cooling modules. In at least one embodiment, one or more of the nodes CR 816(1)-816(N) may be a server having one or more of the computing resources described above.

[0040] In at least one embodiment, the grouped computing resources 814 may include separate groups of node CRs housed in one or more racks (not shown), or multiple racks housed in a data center at various graphical locations (also not shown). The separate groups of node CRs in the grouped computing resources 814 may include grouped compute resources, network resources, memory resources, or storage resources that may be configured or allocated to support one or more workloads. In at least one embodiment, several node CRs including CPUs or processors may be grouped in one or more racks to provide compute resources to support one or more workloads. In at least one embodiment, the one or more racks may also include any number of power modules, cooling modules, and network switches in any combination.

[0041] In at least one embodiment, resource orchestrator 812 may configure or otherwise control one or more nodes CR 816(1)-816(N) and / or grouped computing resources 814. In at least one embodiment, resource orchestrator 812 may include a software design infrastructure ("SDI") management entity for data center 800. In at least one embodiment, resource orchestrator may include hardware, software, or some combination thereof.

[0042] In at least one embodiment, as shown in FIG. 8 , framework layer 820 includes a job scheduler 822, a configuration manager 824, a resource manager 826, and a distributed file system 828. In at least one embodiment, framework layer 820 may include a framework for supporting software 832 in software layer 830 and / or one or more applications 842 in application layer 840. In at least one embodiment, software 832 or applications 842 may each include web-based service software or applications, such as those offered by Amazon Web Services, Google Cloud, and Microsoft Azure. In at least one embodiment, framework layer 820 may be a type of free and open source software web application framework, such as, but not limited to, Apache Spark® (hereinafter “Spark”), which can use distributed file system 828 for large-scale data processing (e.g., “big data”). In at least one embodiment, the job scheduler 822 may include a Spark driver to facilitate scheduling of workloads supported by various tiers of the data center 800. In at least one embodiment, the configuration manager 824 may be capable of configuring different tiers, such as the software tier 830, as well as the framework tier 820 including Spark and a distributed file system 828 to support large scale data processing. In at least one embodiment, the resource manager 826 may be capable of managing clustered or grouped computing resources that are mapped or allocated to support the distributed file system 828 and the job scheduler 822. In at least one embodiment, the clustered or grouped computing resources may include grouped computing resources 814 in the data center infrastructure tier 810.In at least one embodiment, resource manager 826 may cooperate with resource orchestrator 812 to manage these mappings or allocated computing resources.

[0043] In at least one embodiment, software 832 included in software layer 830 may include software used by at least a portion of nodes CR 816(1)-816(N), grouped computing resources 814, and / or distributed file system 828 of framework layer 820. The one or more types of software may include, but are not limited to, Internet web page searching software, email virus scanning software, database software, and streaming video content software.

[0044] In at least one embodiment, the applications 842 included in the application layer 840 may include one or more types of applications used by at least a portion of the nodes CRs 816(1)-816(N), the grouped computing resources 814, and / or the distributed file system 828 of the framework layer 820. The one or more types of applications may include, but are not limited to, any number of genomics applications, cognitive compute, and machine learning applications including training or inference software, machine learning framework software (e.g., PyTorch, TensorFlow, Caffe, etc.), or other machine learning applications used in conjunction with one or more embodiments.

[0045] In at least one embodiment, any of configuration manager 824, resource manager 826, and resource orchestrator 812 may implement any number and type of self-correcting actions based on any amount and type of data obtained in any technically feasible manner. In at least one embodiment, the self-correcting actions may enable data center operators of data center 800 to avoid determining potentially poor configurations and eliminate underutilized and / or underperforming portions of the data center.

[0046] In at least one embodiment, data center 800 may include tools, services, software, or other resources for training one or more machine learning models or predicting or inferring information using one or more machine learning models according to one or more embodiments described herein. For example, in at least one embodiment, a machine learning model may be trained by calculating weight parameters according to a neural network architecture using the software and computing resources described above with respect to data center 800. In at least one embodiment, a trained machine learning model corresponding to one or more neural networks may be used to infer or predict information using the resources described above with respect to data center 800 by using weight parameters calculated by one or more techniques described herein.

[0047] In at least one embodiment, the data center may use a CPU, application specific integrated circuit (ASIC), GPU, FPGA, or other hardware to perform training and / or inference using the resources described above. Additionally, one or more of the software and / or hardware resources described above may be configured as a service to enable a user to train or perform inference on information, such as image recognition, speech recognition, or other artificial intelligence services.

[0048] Inference and / or training logic 715 is used to perform inference and / or training operations associated with one or more embodiments. More details regarding the inference and / or training logic 715 are provided below in conjunction with Figures 7A and / or 7B. In at least one embodiment, the inference and / or training logic 715 may be used in the system of Figure 8 for inference or prediction operations based at least in part on weight parameters calculated using neural network training operations, neural network functionality and / or architecture, or neural network use cases described herein.

[0049] Using these components, live updates of machine learning models can be performed for applications running at locations such as the network edge, without any downtime as the application switches to a different model version, or without any significant downtime.

[0050] Computer Systems 9 is a block diagram illustrating an exemplary computer system, which may be a system having interconnected devices and components formed with a processor that may include an execution unit for executing instructions, a system on a chip (SoC), or some combination of these 900, according to at least one embodiment. In at least one embodiment, computer system 900 may include components such as, without limitation, a processor 902 for using an execution unit that includes logic for executing algorithms for processing data in accordance with the present disclosure, such as in the embodiments described herein. In at least one embodiment, computer system 900 may include a processor such as the PENTIUM® processor family, Xeon™, Itanium®, XScale™ and / or StrongARM™, Intel® Core™, or Intel® Nervana™ microprocessors available from Intel Corporation, Santa Clara, Calif., although other systems may be used (including PCs with other microprocessors, engineering workstations, set-top boxes, etc.). In at least one embodiment, computer system 900 may run a version of the WINDOWS® operating system available from Microsoft Corporation of Redmond, Washington, although other operating systems (e.g., UNIX® and Linux®), embedded software, and / or graphical user interfaces may be used.

[0051] Embodiments may be used in other devices, such as portable devices and embedded applications. Some examples of portable devices include cellular phones, Internet protocol devices, digital cameras, personal digital assistants ("PDAs"), and portable PCs. In at least one embodiment, embedded applications may include microcontrollers, digital signal processors ("DSPs"), systems-on-chips, network computers ("NetPCs"), set-top boxes, network hubs, wide area network ("WAN") switches, or any other system capable of executing one or more instructions according to at least one embodiment.

[0052] In at least one embodiment, computer system 900 may include, without limitation, a processor 902, which may include one or more execution units 908 for performing training and / or inference of machine learning models according to the techniques described herein. In at least one embodiment, computer system 900 is a single processor desktop or server system, while in other embodiments, computer system 900 may be a multiprocessor system. In at least one embodiment, processor 902 may include, without limitation, a complex instruction set computer ("CISC") microprocessor, a reduced instruction set computing ("RISC") microprocessor, a very long instruction word ("VLIW") microprocessor, a processor implementing a combination of instruction sets, or any other processor device, such as a digital signal processor. In at least one embodiment, processor 902 may be coupled to a processor bus 910, which may transmit data signals between processor 902 and other components within computer system 900.

[0053] In at least one embodiment, the processor 902 may include, without limitation, a level 1 ("L1") internal cache memory ("cache") 904. In at least one embodiment, the processor 902 may have a single internal cache or multiple levels of internal cache. In at least one embodiment, the cache memory may be external to the processor 902. Other embodiments may include a combination of both internal and external cache depending on the particular implementation and needs. In at least one embodiment, the register file 906 may store different types of data in various registers, including, without limitation, integer registers, floating point registers, status registers, and an instruction pointer register.

[0054] In at least one embodiment, processor 902 also includes an execution unit 908, including, without limitation, logic for performing integer and floating point operations. In at least one embodiment, processor 902 may also include a microcode ("u-code") read only memory ("ROM") that stores microcode for certain macroinstructions. In at least one embodiment, execution unit 908 may include logic for a packed instruction set 909. In at least one embodiment, by including packed instruction set 909 in the instruction set of a general purpose processor along with associated circuitry for executing the instructions, operations used by many multimedia applications may be performed using packed data in general purpose processor 902. In one or more embodiments, many multimedia applications may be accelerated and executed more efficiently by performing operations on packed data using the full width of the processor's data bus, thereby eliminating the need to transfer smaller units of data between the processor's data bus to perform one or more operations on one data element at a time.

[0055] In at least one embodiment, the execution unit 908 may also be used in a microcontroller, an embedded processor, a graphics device, a DSP, and other types of logic circuits. In at least one embodiment, the computer system 900 may include, without limitation, a memory 920. In at least one embodiment, the memory 920 may be implemented as a dynamic random access memory ("DRAM") device, a static random access memory ("SRAM") device, a flash memory device, or other memory device. In at least one embodiment, the memory 920 may store instructions 919 and / or data 921 represented by data signals that may be executed by the processor 902.

[0056] In at least one embodiment, a system logic chip may be coupled to the processor bus 910 and memory 920. In at least one embodiment, the system logic chip may include, without limitation, a memory controller hub ("MCH") 916, and the processor 902 may communicate with the MCH 916 via the processor bus 910. In at least one embodiment, the MCH 916 may provide a high bandwidth memory path 918 to the memory 920 for storing instructions and data, and for storing graphics commands, data, and textures. In at least one embodiment, the MCH 916 may route data signals between the processor 902, the memory 920, and other components of the computer system 900, and may bridge data signals between the processor bus 910, the memory 920, and a system I / O interface 922. In at least one embodiment, the system logic chip may provide a graphics port for coupling to a graphics controller. In at least one embodiment, the MCH 916 may be coupled to memory 920 via a high bandwidth memory path 918, and the graphics / video card 912 may be coupled to the MCH 916 via an Accelerated Graphics Port ("AGP") interconnect 914.

[0057] In at least one embodiment, computer system 900 may use system I / O 922, which is a proprietary hub interface bus for coupling MCH 916 to I / O controller hub ("ICH") 930. In at least one embodiment, ICH 930 may provide direct connectivity to some I / O devices via a local I / O bus. In at least one embodiment, local I / O bus may include, without limitation, a high-speed I / O bus for connecting peripherals to memory 920, chipset, and processor 902. Examples may include, without limitation, an audio controller 929, a firmware hub ("flash BIOS") 928, wireless transceiver 926, data storage 924, legacy I / O controller 923 including user input and keyboard interface 925, serial expansion port 927 such as Universal Serial Bus ("USB"), and network controller 934. Data storage 924 may include a hard disk drive, a floppy disk drive, a CD-ROM device, a flash memory device, or other mass storage device.

[0058] In at least one embodiment, Figure 9 illustrates a system including interconnected hardware devices or "chips," while in other embodiments, Figure 9 may illustrate an exemplary system-on-a-chip ("SoC"). In at least one embodiment, the devices may be interconnected with proprietary interconnects, standard interconnects (e.g., PCIe), or some combination thereof. In at least one embodiment, one or more components of computer system 900 may be interconnected using a compute express link (CXL) interconnect.

[0059] Inference and / or training logic 715 is used to perform inference and / or training operations associated with one or more embodiments. More details regarding inference and / or training logic 715 are provided below in conjunction with Figures 7A and / or 7B. In at least one embodiment, inference and / or training logic 715 may be used in the system of Figure 9 for inference or prediction operations based at least in part on weight parameters calculated using neural network training operations, neural network functionality and / or architecture, or neural network use cases described herein.

[0060] Using these components, live updates of machine learning models can be performed for applications running at locations such as the network edge, without any downtime as the application switches to a different model version, or without any significant downtime.

[0061] 10 is a block diagram illustrating an electronic device 1000 for utilizing a processor 1010, according to at least one embodiment. In at least one embodiment, electronic device 1000 may be, for example, without limitation, a notebook, a tower server, a rack server, a blade server, a laptop, a desktop, a tablet, a mobile device, a phone, an embedded computer, or any other suitable electronic device.

[0062] In at least one embodiment, system 1000 may include a processor 1010 communicatively coupled to any suitable number or type of components, peripherals, modules, or devices, without limitation. In at least one embodiment, processor 1010 is coupled using a bus or interface, such as an I°C bus, a System Management Bus (“SMBus”), a Low Pin Count (LPC) bus, a Serial Peripheral Interface (“SPI”), a High Definition Audio (“HDA”) bus, a Serial Advance Technology Attachment (“SATA”) bus, a Universal Serial Bus (“USB”) (versions 1, 2, 3), or a Universal Asynchronous Receiver / Transmitter (“UART”) bus. In at least one embodiment, FIG. 10 illustrates a system including interconnected hardware devices or "chips," while in other embodiments, FIG. 10 may illustrate an exemplary system-on-a-chip ("SoC"). In at least one embodiment, the devices illustrated in FIG. 10 may be interconnected with proprietary interconnects, standard interconnects (e.g., PCIe), or some combination thereof. In at least one embodiment, one or more components of FIG. 10 may be interconnected using a Compute Express Link (CXL) interconnect.

[0063] In at least one embodiment, FIG. 10 illustrates a display 1024, a touch screen 1025, a touch pad 1030, a Near Field Communications unit ("NFC") 1045, a sensor hub 1040, a thermal sensor 1046, an Express Chipset ("EC") 1035, a Trusted Platform Module ("TPM") 1038, a BIOS / firmware / flash memory ("BIOS,FW flash") 1022, a DSP 1060, a drive 1020, such as a Solid State Disk ("SSD") or a Hard Disk Drive ("HDD"), a wireless local area network unit ("WLAN") 1050, a Bluetooth unit 1052, a Wireless Wide Area Network unit ("WWAN") 1054, a Bluetooth 1056, a Bluetooth 2000 (BLE) 1058, a Bluetooth 2002 (BLE) 1059, a Bluetooth 2004 (BLE) 1060, a Bluetooth 2006 (BLE) 1070, a Bluetooth 2008 (BLE) 1072, a Bluetooth 2009 (BLE) 1080, a Bluetooth 2009 (BLE) 1090, a Bluetooth 2001 (BLE) 1092, a Bluetooth 2002 (BLE) 1094, a Bluetooth 2003 (BLE) 1096, a Bluetooth 2004 (BLE) 1098, a Bluetooth 2005 (BLE) 1099, a Bluetooth 2006 (BLE) 1091, a Bluetooth 2007 (BLE) 1092, a Bluetooth 2008 (BLE) 1091, a Bluetooth 2009 (BLE) 1092, a Bluetooth 2009 (BLE) 1094, a Bluetooth 2009 (BLE) 1095, a Bluetooth 2 The memory may include a GPS (Global Positioning System) unit 1056, a Global Positioning System (GPS) unit 1055, a camera, such as a USB 3.0 camera ("USB 3.0 Camera") 1054, and / or a Low Power Double Data Rate ("LPDDR") memory unit ("LPDDR3") 1015, implemented, for example, to the LPDDR3 standard. Each of these components may be implemented in any suitable manner.

[0064] In at least one embodiment, other components may be communicatively coupled to the processor 1010 via the components described above. In at least one embodiment, an accelerometer 1041, an ambient light sensor ("ALS") 1042, a compass 1043, and a gyroscope 1044 may be communicatively coupled to the sensor hub 1040. In at least one embodiment, a thermal sensor 1039, a fan 1037, a keyboard 1046, and a touch pad 1030 may be communicatively coupled to the EC 1035. In at least one embodiment, a speaker 1063, a headphone 1064, and a microphone ("mic") 1065 may be communicatively coupled to an audio unit (audio codec and class d amplifier) ​​1062, which may be communicatively coupled to the DSP 1060. In at least one embodiment, the audio unit 1064 may include, for example, without limitation, an audio coder / decoder ("codec") and a class D amplifier. In at least one embodiment, a SIM card ("SIM") 1057 may be communicatively coupled to the WWAN unit 1056. In at least one embodiment, components such as the WLAN unit 1050 and the Bluetooth unit 1052, as well as the WWAN 1056, may be implemented in a Next Generation Form Factor ("NGFF").

[0065] Inference and / or training logic 715 is used to perform inference and / or training operations associated with one or more embodiments. More details regarding inference and / or training logic 715 are provided below in conjunction with Figures 7A and / or 7B. In at least one embodiment, inference and / or training logic 715 may be used in the system of Figure 10 for inference or prediction operations based at least in part on weight parameters calculated using neural network training operations, neural network functionality and / or architecture, or neural network use cases described herein.

[0066] Using these components, live updates of machine learning models can be performed for applications running at locations such as the network edge, without any downtime as the application switches to a different model version, or without any significant downtime.

[0067] 11 is a block diagram of at least one example processing system. In at least one embodiment, system 1100 includes one or more processors 1102 and one or more graphics processors 1108 and may be a single processor desktop system, a multiprocessor workstation system, or a server system having multiple processors 1102 or processor cores 1107. In at least one embodiment, system 1100 is a processing platform integrated into a system-on-chip (SoC) integrated circuit for use in a mobile, portable, or embedded device.

[0068] In at least one embodiment, the system 1100 may include or be incorporated into a server-based gaming platform, a game console including a game and media console, a mobile gaming console, a handheld game console, or an online game console. In at least one embodiment, the system 1100 is a mobile phone, a smart phone, a tablet computing device, or a mobile Internet device. In at least one embodiment, the processing system 1100 may also include, be coupled to, or be integrated into a wearable device, such as a smart watch wearable device, a smart eyewear device, an augmented reality device, or a virtual reality device. In at least one embodiment, the processing system 1100 is a television or set-top box device having one or more processors 1102 and a graphical interface generated by one or more graphics processors 1108.

[0069] In at least one embodiment, the one or more processors 1102 each include one or more processor cores 1107 for processing instructions that, when executed, perform operations for system and user software. In at least one embodiment, each of the one or more processor cores 1107 is configured to process a particular instruction set 1109. In at least one embodiment, the instruction set 1109 may facilitate complex instruction set computing (CISC), reduced instruction set computing (RISC), or computing via very long instruction words (VLIW). In at least one embodiment, the processor cores 1107 each may process a different instruction set 1109, which may include instructions that facilitate emulation of other instruction sets. In at least one embodiment, the processor cores 1107 may also include other processing devices, such as digital signal processors (DSPs).

[0070] In at least one embodiment, the processor 1102 includes a cache memory 1104. In at least one embodiment, the processor 1102 may have a single internal cache or multiple levels of internal cache. In at least one embodiment, the cache memory is shared among various components of the processor 1102. In at least one embodiment, the processor 1102 also uses an external cache (e.g., a level 3 (L3) cache or a last level cache (LLC)) (not shown), which may be shared among the processor cores 1107 using known cache coherence techniques. In at least one embodiment, the processor 1102 further includes a register file 1106, which may include different types of registers (e.g., integer registers, floating point registers, status registers, and an instruction pointer register) for storing different types of data. In at least one embodiment, the register file 1106 may include general purpose registers or other registers.

[0071] In at least one embodiment, the one or more processors 1102 are coupled to one or more interface buses 1110 to transmit communication signals, such as address, data, or control signals, between the processors 1102 and other components in the system 1100. In at least one embodiment, the interface bus 1110 may be a processor bus, such as, in one embodiment, a version of a Direct Media Interface (DMI) bus. In at least one embodiment, the interface 1110 is not limited to a DMI bus, but may include one or more peripheral component interconnect buses (e.g., PCI, PCI Express), a memory bus, or other types of interface buses. In at least one embodiment, the processor 1102 includes an integrated memory controller 1116 and a platform controller hub 1130. In at least one embodiment, the memory controller 1116 facilitates communication between memory devices and other components of the system 1100, while the platform controller hub (PCH) 1130 provides connectivity to I / O devices via a local I / O bus.

[0072] In at least one embodiment, the memory device 1120 may be a dynamic random access memory (DRAM) device, a static random access memory (SRAM) device, a flash memory device, a phase change memory device, or any other memory device having suitable performance to serve as process memory. In at least one embodiment, the memory device 1120 may operate as system memory for the system 1100 to store data 1122 and instructions 1121 for use when the one or more processors 1102 execute applications or processes. In at least one embodiment, the memory controller 1116 is also coupled to an optional external graphics processor 1112, which may communicate with one or more graphics processors 1108 within the processor 1102 to perform graphics and media operations. In at least one embodiment, a display device 1111 may be connected to the processor 1102. In at least one embodiment, the display device 1111 may include one or more of an internal display device, such as a mobile electronic device or laptop device, or an external display device attached via a display interface (e.g., a display port, etc.). In at least one embodiment, the display device 1111 may include a head mounted display (HMD), such as a stereoscopic display device for use in virtual reality (VR) or augmented reality (AR) applications.

[0073] In at least one embodiment, the platform controller hub 1130 allows peripherals to connect to the memory device 1120 and the processor 1102 via a high-speed I / O bus. In at least one embodiment, the I / O peripherals include, but are not limited to, an audio controller 1146, a network controller 1134, a firmware interface 1128, a wireless transceiver 1126, a touch sensor 1125, and a data storage device 1124 (e.g., hard disk drive, flash memory, etc.). In at least one embodiment, the data storage device 1124 can be connected via a storage interface (e.g., SATA) or via a peripheral bus such as a peripheral component interconnect bus (e.g., PCI, PCI Express). In at least one embodiment, the touch sensor 1125 can include a touch screen sensor, a pressure sensor, or a fingerprint sensor. In at least one embodiment, the wireless transceiver 1126 may be a WiFi transceiver, a Bluetooth transceiver, or a mobile network transceiver such as a 3G, 4G, or Long Term Evolution (LTE) transceiver. In at least one embodiment, the firmware interface 1128 enables communication with system firmware and may be, for example, a Unified Extensible Firmware Interface (UEFI). In at least one embodiment, the network controller 1134 may enable network connectivity to a wired network. In at least one embodiment, a high performance network controller (not shown) couples to the interface bus 1110. In at least one embodiment, the audio controller 1146 is a multi-channel high definition audio controller. In at least one embodiment, the system 1100 includes an optional legacy I / O controller 1140 for coupling legacy (e.g., Personal System 2 (PS / 2)) devices to the system.In at least one embodiment, platform controller hub 1130 may also connect to one or more universal serial bus (USB) controllers 1142 for connecting input devices, such as a keyboard and mouse 1143 combination, a camera 1144, or other USB input devices.

[0074] In at least one embodiment, instances of memory controller 1116 and platform controller hub 1130 may be integrated into a separate external graphics processor, such as external graphics processor 1112. In at least one embodiment, platform controller hub 1130 and / or memory controller 1116 may be external to one or more processors 1102. For example, in at least one embodiment, system 1100 may include an external memory controller 1116 and platform controller hub 1130, which may be configured as a memory controller hub and a peripheral controller hub in a system chipset that communicates with processor 1102.

[0075] Inference and / or training logic 715 is used to perform inference and / or training operations associated with one or more embodiments. More details regarding the inference and / or training logic 715 are provided below in conjunction with FIG. 7A and / or FIG. 7B. In at least one embodiment, some or all of the inference and / or training logic 715 may be incorporated into the graphics processor 1500. For example, in at least one embodiment, the training and / or inference techniques described herein may use one or more of the ALUs embodied in the graphics processor. Furthermore, in at least one embodiment, the inference and / or training operations described herein may be performed using logic other than that shown in FIG. 7A or FIG. 7B. In at least one embodiment, the weight parameters may be stored in on-chip or off-chip memory and / or registers (as shown or not shown) that configure the ALUs of the graphics processor to execute one or more machine learning algorithms, neural network architectures, use cases, or training techniques described herein.

[0076] Using these components, live updates of machine learning models can be performed for applications running at locations such as the network edge, without any downtime as the application switches to a different model version, or without any significant downtime.

[0077] 12 is a block diagram of a processor 1200 having one or more processor cores 1202A-1202N, an integrated memory controller 1214, and an integrated graphics processor 1208, according to at least one embodiment. In at least one embodiment, the processor 1200 may include a smaller number of additional cores, including additional core 1202N, represented by a dashed box. In at least one embodiment, each of the processor cores 1202A-1202N includes one or more internal cache units 1204A-1204N. In at least one embodiment, each processor core also has access to one or more shared cache units 1206.

[0078] In at least one embodiment, the internal cache units 1204A-1204N and the shared cache unit 1206 represent a cache memory hierarchy within the processor 1200. In at least one embodiment, the cache memory units 1204A-1204N may include at least one level of instruction and data cache within each processor core, as well as one or more levels of shared mid-level cache, such as level 2 (L2), level 3 (L3), level 4 (L4), or other levels of cache, where the highest level of cache before external memory is classified as an LLC. In at least one embodiment, cache coherence logic maintains coherency between the various cache units 1206 and 1204A-1204N.

[0079] In at least one embodiment, processor 1200 may also include a set of one or more bus controller units 1216 and a system agent core 1210. In at least one embodiment, one or more bus controller units 1216 manage a set of peripheral buses, such as one or more PCI or PCI Express buses. In at least one embodiment, system agent core 1210 provides management functions for various processor components. In at least one embodiment, system agent core 1210 includes one or more integrated memory controllers 1214 for managing access to various external memory devices (not shown).

[0080] In at least one embodiment, one or more of the processor cores 1202A-1202N include support for simultaneous multithreading. In at least one embodiment, the system agent core 1210 includes components for coordinating and operating the cores 1202A-1202N during multithreaded processing. In at least one embodiment, the system agent core 1210 may further include a power control unit (PCU), which includes logic and components for coordinating the power state of one or more of the processor cores 1202A-1202N and the graphics processor 1208.

[0081] In at least one embodiment, processor 1200 further includes a graphics processor 1208 for performing graphics processing operations. In at least one embodiment, graphics processor 1208 couples to shared cache unit 1206 and to a system agent core 1210 that includes one or more integrated memory controllers 1214. In at least one embodiment, system agent core 1210 also includes a display controller 1211 for directing output of the graphics processor to one or more coupled displays. In at least one embodiment, display controller 1211 may also be a separate module coupled to graphics processor 1208 via at least one interconnect or may be integrated within graphics processor 1208.

[0082] In at least one embodiment, a ring-based interconnect unit 1212 is used to couple the internal components of processor 1200. In at least one embodiment, alternative interconnect units such as a point-to-point interconnect, a switched interconnect, or other techniques may be used. In at least one embodiment, graphics processor 1208 couples to ring interconnect 1212 via I / O link 1213.

[0083] In at least one embodiment, I / O link 1213 represents at least one of a variety of I / O interconnects, including an on-package I / O interconnect that facilitates communication between various processor components and a high performance embedded memory module 1218, such as an eDRAM module. In at least one embodiment, each of processor cores 1202A-1202N and graphics processor 1208 use embedded memory module 1218 as a shared last level cache.

[0084] In at least one embodiment, the processor cores 1202A-1202N are homogenous cores that execute a common instruction set architecture. In at least one embodiment, the processor cores 1202A-1202N are heterogeneous in terms of instruction set architecture (ISA), where one or more of the processor cores 1202A-1202N execute a common instruction set, while one or more other of the processor cores 1202A-1202N execute a subset of the common instruction set, or a different instruction set. In at least one embodiment, the processor cores 1202A-1202N are heterogeneous in terms of micro-architecture, where one or more cores with relatively higher power consumption are combined with one or more cores with lower power consumption. In at least one embodiment, the processor 1200 can be implemented on one or more chips or as an SoC integrated circuit.

[0085] Inference and / or training logic 715 is used to perform inference and / or training operations associated with one or more embodiments. More details regarding inference and / or training logic 715 are provided below in conjunction with FIG. 7A and / or FIG. 7B. In at least one embodiment, some or all of the inference and / or training logic 715 may be incorporated into processor 1200. For example, in at least one embodiment, the training and / or inference techniques described herein may use one or more of the ALUs embodied in graphics processor 1512, graphics cores 1202A-1202N, or other components of FIG. 12. Furthermore, in at least one embodiment, the inference and / or training operations described herein may be performed using logic other than that shown in FIG. 7A or FIG. 7B. In at least one embodiment, the weight parameters may be stored in on-chip or off-chip memory and / or registers (as shown or not shown) that configure the ALUs of graphics processor 1200 to perform one or more machine learning algorithms, neural network architectures, use cases, or training techniques described herein.

[0086] Using these components, live updates of machine learning models can be performed for applications running at locations such as the network edge, without any downtime as the application switches to a different model version, or without any significant downtime.

[0087] Virtualized Computing Platform FIG. 13 is an example data flow diagram of a process 1300 for generating and deploying an image processing and inference pipeline, according to at least one embodiment. In at least one embodiment, the process 1300 may be deployed for use with imaging devices, processing devices, and / or other types of devices at one or more facilities 1302. In at least one embodiment, the process 1300 may be performed in a training system 1304 and / or in a deployment system 1306. In at least one embodiment, the training system 1304 may be used to train, deploy, and implement machine learning models (e.g., neural networks, object detection algorithms, computer vision algorithms, etc.) for use in the deployment system 1306. In at least one embodiment, the deployment system 1306 may be configured to offload processing and computational resources between distributed computing environments to reduce infrastructure requirements at the facility 1302. In at least one embodiment, one or more applications in the pipeline may use or call services (e.g., inference, virtualization, computation, AI, etc.) of the deployment system 1306 during execution of the applications.

[0088] In at least one embodiment, some of the applications used in the advanced processing and inference pipeline may use machine learning models or other AI to perform one or more processing steps. In at least one embodiment, the machine learning models may be trained at the facility 1302 using data (e.g., imaging data) 1308 generated at the facility 1302 (and stored in one or more picture archiving and communication system (PACS) servers at the facility 1302), may be trained using imaging or sequencing data 1308 from one or more other facilities, or a combination thereof. In at least one embodiment, a training system 1304 may be used to provide applications, services, and / or other resources for generating practical, deployable machine learning models for a deployment system 1306.

[0089] In at least one embodiment, the model registry 1324 may be backed by an object storage that may support versioning and object metadata. In at least one embodiment, the object storage may be accessible, for example, from within a cloud platform via a cloud storage (e.g., cloud 1426 of FIG. 14) compatibility application programming interface (API). In at least one embodiment, machine learning models in the model registry 1324 may be uploaded, listed, modified, or deleted by a system developer or partner interacting with the API. In at least one embodiment, the API may provide access to a method that allows a user with appropriate credentials to associate a model with an application, thereby allowing the model to be executed as part of running a containerized instance of the application.

[0090] In at least one embodiment, the training pipeline 1404 (FIG. 14) may include a situation where the facility 1302 is training its own machine learning model or has an existing machine learning model that needs to be optimized or updated. In at least one embodiment, imaging data 1308 generated by an imaging device, a sequencing device, and / or other type of device may be received. In at least one embodiment, once the imaging data 1308 is received, AI-assisted annotation 1310 may be used to assist in generating annotations corresponding to the imaging data 1308 that will be used as ground truth data for the machine learning model. In at least one embodiment, the AI-assisted annotation 1310 may include one or more machine learning models (e.g., convolutional neural networks (CNNs)), which may be trained to generate annotations corresponding to a particular type of imaging data 1308 (e.g., from a particular device). In at least one embodiment, the AI-assisted annotation 1310 may then be used directly to generate ground truth data or may be adjusted or fine-tuned using annotation tools. In at least one embodiment, the AI-assisted annotations 1310, the labeled clinic data 1312, or a combination thereof may be used as ground truth data to train a machine learning model. In at least one embodiment, the trained machine learning model may be referred to as an output model 1316 and may be used by the deployment system 1306 described herein.

[0091] In at least one embodiment, the training pipeline 1404 (FIG. 14) may include a situation where the facility 1302 needs a machine learning model to use in performing one or more processing tasks for one or more applications in the installation system 1306, but the facility 1302 may not currently have such a machine learning model (or may not have a model that is optimized, efficient, or effective for such purposes). In at least one embodiment, an existing machine learning model may be selected from the model registry 1324. In at least one embodiment, the model registry 1324 may include machine learning models that have been trained to perform a variety of different inference tasks on imaging data. In at least one embodiment, the machine learning models in the model registry 1324 may have been trained on imaging data from a facility different from the facility 1302 (e.g., a facility in a remote location). In at least one embodiment, the machine learning models may have been trained on imaging data from one location, two locations, or any number of locations. In at least one embodiment, when trained on imaging data from a particular location, the training may be performed at the location, or at least in a manner that protects the confidentiality of the imaging data or limits the imaging data from being transferred off-premise. In at least one embodiment, once a model is trained or partially trained at one location, the machine learning model may be added to a model registry 1324. In at least one embodiment, the machine learning model may then be retrained or updated at any number of other facilities, and the retrained or updated model may be made available in the model registry 1324. In at least one embodiment, the machine learning model may then be selected from the model registry 1324, and may be referred to as an output model 1316, and may be used in the installation system 1306 to perform one or more processing tasks for one or more applications of the installation system.

[0092] In at least one embodiment, the training pipeline 1404 (FIG. 14) may include a scenario in which the facility 1302 needs a machine learning model to use in performing one or more processing tasks for one or more applications in the deployment system 1306, but the facility 1302 may not currently have such a machine learning model (or may not have a model optimized, efficient, or effective for such purposes). In at least one embodiment, the machine learning model selected from the model registry 1324 may not be fine-tuned or optimized for the imaging data 1308 generated at the facility 1302 due to differences in the population, robustness of the training data used to train the machine learning model, variety of anomalies in the training data, and / or other issues with the training data. In at least one embodiment, AI-assisted annotation 1310 may be used to assist in generating annotations corresponding to the imaging data 1308 to be used as ground truth data for retraining or updating the machine learning model. In at least one embodiment, the labeled data 1312 may be used as ground truth data for training the machine learning model. In at least one embodiment, retraining or updating the machine learning model may be referred to as model training 1314. In at least one embodiment, the model training 1314, e.g., the AI-assisted annotations 1310, the labeled clinic data 1312, or a combination thereof, may be used as ground truth data to retrain or update the machine learning model. In at least one embodiment, the trained machine learning model may be referred to as output model 1316 and may be used by the deployment system 1306 described herein.

[0093] In at least one embodiment, the deployment system 1306 may include software 1318, services 1320, hardware 1322, and / or other components, features, and functions. In at least one embodiment, the deployment system 1306 may include a software “stack” whereby the software 1318 may be built on top of the services 1320 and may use the services 1320 to perform some or all of the processing tasks, and the services 1320 and the software 1318 may be built on top of the hardware 1322 and may use the hardware 1322 to perform the processing, storage, and / or other computational tasks of the deployment system 1306. In at least one embodiment, the software 1318 may include any number of different containers, where each container may perform an instantiation of an application. In at least one embodiment, each application may perform one or more processing tasks (e.g., inference, object detection, feature detection, segmentation, image enhancement, calibration, etc.) of an advanced processing and inference pipeline. In at least one embodiment, an advanced processing and inference pipeline may be defined based on a selection of different containers desired or required to process the imaging data 1308, in addition to the containers that receive and compose the imaging data used by each container and / or used by facility 1302 after processing through the pipeline (e.g., to reconvert the output into a usable type of data). In at least one embodiment, the combination of containers in software 1318 (e.g., comprising a pipeline) may be referred to as a virtual device (described in more detail herein), which may utilize services 1320 and hardware 1322 to perform some or all of the processing tasks of the applications instantiated in the containers.

[0094] In at least one embodiment, the data processing pipeline may receive input data (e.g., imaging data 1308) in a particular format in response to an inference request (e.g., a request from a user of the deployment system 1306). In at least one embodiment, the input data may represent one or more images, videos, and / or other data representations generated by one or more imaging devices. In at least one embodiment, the data may undergo pre-processing as part of the data processing pipeline to prepare the data for processing by one or more applications. In at least one embodiment, post-processing may be performed on the output of one or more inference or other processing tasks of the pipeline to prepare output data for a subsequent application and / or for transmission and / or use by a user (e.g., in response to an inference request). In at least one embodiment, the inference tasks may be performed by one or more machine learning models, such as trained or deployed neural networks, which may include the output model 1316 of the training system 1304.

[0095] In at least one embodiment, tasks in a data processing pipeline may be encapsulated in containers, each of which represents a separate, fully functional instantiation of an application and a virtualized computing environment that can reference a machine learning model. In at least one embodiment, containers or applications may be published to a private (e.g., restricted access) area of ​​a container registry (described in more detail herein), and trained or deployed models may be stored in a model registry 1324 and associated with one or more applications. In at least one embodiment, images of the applications (e.g., images of containers) may be available in the container registry, and once selected from the container registry by a user for deployment in the pipeline, the images may be used to generate a container for instantiating the application for use on the user's system.

[0096] In at least one embodiment, a developer (e.g., a software developer, clinician, doctor, etc.) may develop, publish, and store an application (e.g., as a container) to perform image processing and / or inference on the provided data. In at least one embodiment, the development, publishing, and / or storage may be performed using a software development kit (SDK) associated with the system (e.g., to ensure that the developed application and / or container is compliant with or compatible with the system). In at least one embodiment, the developed application may be tested locally (e.g., at the first facility, on data from the first facility) using the SDK that can support at least a portion of the services 1320 as a system (e.g., system 1400 of FIG. 14). In at least one embodiment, because a DICOM object can contain anywhere from one to hundreds of images or other types of data, and there is variation in the data, the developer may be responsible for managing the extraction and preparation of the incoming data (e.g., setting configurations for the application, building pre-processing into the application, etc.). In at least one embodiment, once verified (e.g., for accuracy) by system 1400, the application may be made available in a container registry for selection and / or implementation by a user, and one or more processing tasks may be performed on the data at the user's facility (e.g., a second facility).

[0097] In at least one embodiment, the developer may then share the application or container over a network for access and use by users of the system (e.g., system 1400 of FIG. 14). In at least one embodiment, the completed and validated application or container may be stored in a container registry, and the associated machine learning model may be stored in a model registry 1324. In at least one embodiment, a requesting entity issuing an inference or image processing request may browse the container registry and / or model registry 1324 to find applications, containers, datasets, machine learning models, etc., select a desired combination of elements for inclusion in a data processing pipeline, and submit an image processing request. In at least one embodiment, the request may include input data (and in some instances, associated patient data) required to execute the request, and / or may include a selection of an application and / or machine learning model to be executed in processing the request. In at least one embodiment, the request may then be passed to one or more components of the deployment system 1306 (e.g., cloud) to execute the processing of the data processing pipeline. In at least one embodiment, processing by the deployment system 1306 may include referencing selected elements (e.g., applications, containers, models, etc.) from the container registry and / or the model registry 1324. In at least one embodiment, once results are produced by the pipeline, the results may be returned to a user for viewing (e.g., viewed locally in a viewing application suite running on a workstation or terminal on premises).

[0098] In at least one embodiment, services 1320 may be utilized to assist in the processing or execution of applications or containers in the pipeline. In at least one embodiment, services 1320 may include computational services, artificial intelligence (AI) services, visualization services, and / or other types of services. In at least one embodiment, services 1320 may provide common functionality to one or more applications of software 1318, whereby functionality may be abstracted into services that may be called or utilized by the applications. In at least one embodiment, functionality provided by services 1320 may be dynamically and more efficiently performed while at the same time scaling well by allowing applications to process data in parallel (e.g., using parallel computing platform 1430 (FIG. 14)). In at least one embodiment, services 1320 may be shared among various applications, rather than requiring each application that shares the same functionality provided by services 1320 to have a respective instance of services 1320. In at least one embodiment, services may include, by way of non-limiting example, inference servers or engines that may be used to perform detection or segmentation tasks. At least one embodiment may include a model training service that may provide training and / or retraining of machine learning models. At least one embodiment may further include a data augmentation service that may provide GPU accelerated data (e.g., DICOM, RIS, CIS, REST-compliant, RPC, raw, etc.) extraction, resizing, scaling, and / or other enhancements. At least one embodiment may include a visualization service that may add image rendering effects such as ray tracing, rasterization, denoising, sharpening, etc. to add realism to two-dimensional (2D) and / or three-dimensional (3D) models. At least one embodiment may include a virtual device service that provides beamforming, segmentation, inference, imaging, and / or support for other applications in the virtual device pipeline.

[0099] In at least one embodiment, if the services 1320 include an AI service (e.g., an inference service), the one or more machine learning models may be executed by calling (as an API call) to an inference service (e.g., an inference server) to execute the machine learning models, or processing thereof, as part of the application execution. In at least one embodiment, if another application includes one or more machine learning models for a segmentation task, the application may call the inference service to execute the machine learning models to perform one or more of the processing operations associated with the segmentation task. In at least one embodiment, software 1318 implementing advanced processing and inference pipelines, including a segmentation application and an anomaly detection application, may be streamlined since each application may call the same inference service to perform one or more inference tasks.

[0100] In at least one embodiment, the hardware 1322 may include a GPU, a CPU, a graphics card, an AI / deep learning system (e.g., an AI supercomputer such as NVIDIA's DGX), a cloud platform, or a combination thereof. In at least one embodiment, different types of hardware 1322 may be used to provide efficient and dedicated support for the software 1318 and services 1320 of the deployment system 1306. In at least one embodiment, the use of GPU processing to perform processing locally (e.g., at the facility 1302), within the AI / deep learning system, in the cloud system, and / or in other processing components of the deployment system 1306 to improve the efficiency, accuracy, and effectiveness of image processing. In at least one embodiment, the software 1318 and / or services 1320 may be optimized for GPU processing for deep learning, machine learning, and / or high performance computing, as non-limiting examples. In at least one embodiment, at least a portion of the computing environment of the deployment system 1306 and / or the training system 1304 may be executed in one or more supercomputers or high performance computing systems in a data center using GPU optimization software (e.g., a hardware and software combination of NVIDIA's DGX system). In at least one embodiment, the hardware 1322 may include any number of GPUs, which may be called upon to perform parallel processing of data as described herein. In at least one embodiment, the cloud platform may further include GPU processing for GPU-optimized execution of deep learning tasks, machine learning tasks, or other computing tasks. In at least one embodiment, the cloud platform (e.g., NVIDIA's NGC) may be executed using an AI / deep learning supercomputer (e.g., provided by NVIDIA's DGX system) and / or GPU optimization software as a hardware abstraction and scaling platform.In at least one embodiment, the cloud platform may integrate an application container clustering system or orchestration system (e.g., KUBERNETES) across multiple GPUs to enable seamless scaling and load balancing.

[0101] 14 is a system diagram illustrating an example system 1400 for generating and deploying an imaging deployment pipeline according to at least one embodiment. In at least one embodiment, system 1400 may be used to implement process 1300 of FIG. 13 and / or other processes including advanced processing and inference pipelines. In at least one embodiment, system 1400 may include training system 1304 and deployment system 1306. In at least one embodiment, training system 1304 and deployment system 1306 may be implemented using software 1318, services 1320, and / or hardware 1322 as described herein.

[0102] In at least one embodiment, the system 1400 (e.g., the training system 1304 and / or the deployment system 1306) may be implemented in a cloud computing environment (e.g., the cloud 1426). In at least one embodiment, the system 1400 may be implemented locally with respect to a healthcare service facility or as a combination of cloud and local computing resources. In at least one embodiment, access to the APIs of the cloud 1426 may be limited to authorized users via enacted security measures or protocols. In at least one embodiment, the security protocol may include a web token, which may be signed by an authentication (e.g., AuthN, AuthZ, Gluecon, etc.) service and may have appropriate permissions. In at least one embodiment, the APIs of a virtual device (as described herein) or other instantiation of the system 1400 may be limited to a set of public IPs that have been verified or authorized for interaction.

[0103] In at least one embodiment, the various components of system 1400 may communicate with one another using any of a variety of different types of networks, including, but not limited to, a local area network (LAN) and / or a wide area network (WAN), via wired and / or wireless communication protocols. In at least one embodiment, communications between the facility and the components of system 1400 (e.g., for sending inference requests, receiving results of inference requests, etc.) may be communicated via one or more data buses, wireless data protocols (Wi-Fi), wired data protocols (e.g., Ethernet), etc.

[0104] In at least one embodiment, the training system 1304 may execute a training pipeline 1404 similar to that described herein with respect to FIG. 13. In at least one embodiment, if one or more machine learning models are to be used by the deployment system 1306 in the deployment pipeline 1410, the training pipeline 1404 may be used to train or retrain one or more (e.g., pre-trained) models and / or to implement one or more of the pre-trained models 1406 (e.g., without the need for retraining or updating). In at least one embodiment, the training pipeline 1404 may result in an output model 1316. In at least one embodiment, the training pipeline 1404 may include any number of processing steps, such as, but not limited to, transforming or adapting the imaging data (or other input data). In at least one embodiment, different training pipelines 1404 may be used for different machine learning models used by the deployment system 1306. In at least one embodiment, a training pipeline 1404 similar to the first example described with respect to FIG. 13 may be used for the first machine learning model, a training pipeline 1404 similar to the second example described with respect to FIG. 13 may be used for the second machine learning model, and a training pipeline 1404 similar to the third example described with respect to FIG. 13 may be used for the third machine learning model. In at least one embodiment, any combination of tasks in the training system 1304 may be used depending on what is required for each respective machine learning model. In at least one embodiment, one or more of the machine learning models may already be trained and ready for deployment, such that the machine learning models do not need to undergo any processing by the training system 1304 and may be implemented by the deployment system 1306.

[0105] In at least one embodiment, the output model 1316 and / or the pre-trained model 1406 may include any type of machine learning model depending on the implementation or embodiment. In at least one embodiment, without limitation, the machine learning models used by the system 1400 may include machine learning models using linear regression, logistic regression, decision trees, support vector machines (SVMs), naive Bayes, k-nearest neighbor (Knn), k-means clustering, random forests, dimensionality reduction algorithms, gradient boosting algorithms, neural networks (e.g., autoencoders, convolutional, recurrent, perceptrons, long / short-term memory (LSTM), Hopfield, Boltzmann, deep belief, deconvolution, generative adversarial, liquid state machines, etc.), and / or other types of machine learning models.

[0106] In at least one embodiment, the training pipeline 1404 may include AI-assisted annotations, as described in more detail herein with respect to at least FIG. 15B. In at least one embodiment, the labeled clinic data 1312 (e.g., traditional annotations) may be generated by any number of techniques. In at least one embodiment, the labels or other annotations may be generated within a drawing program (e.g., an annotation program), a computer-aided design (CAD) program, a labeling program, another type of program suitable for generating annotations or labels for ground truthing, and / or may be hand-drawn in some instances. In at least one embodiment, the ground truth data may be synthetically generated (e.g., generated from a computer model or rendering), realistically generated (e.g., designed and generated from real-world data), machine automated (e.g., using feature analysis and learning to extract features from data and then generate labels), human annotated (e.g., a labeler or annotation expert defines the location of the labels), and / or a combination thereof. In at least one embodiment, for each instance of imaging data 1308 (or other type of data used by a machine learning model), there may be corresponding ground truth data generated by the training system 1304. In at least one embodiment, AI-assisted annotation may be performed as part of the deployment pipeline 1410 in addition to or instead of AI-assisted annotation included in the training pipeline 1404. In at least one embodiment, system 1400 may include a multi-tier platform that may include a software layer (e.g., software 1318) of diagnostic applications (or other types of applications) that may perform one or more medical imaging and diagnostic functions. In at least one embodiment, system 1400 may be communicatively coupled (e.g., via an encrypted link) to one or more facility's PACS server networks.In at least one embodiment, system 1400 may be configured to access and reference data from a PACS server to perform operations such as training machine learning models, deploying machine learning models, image processing, inference, and / or other operations.

[0107] In at least one embodiment, the software layer may be implemented as a secure, encrypted, and / or authenticated API through which an application or container may be invoked (e.g., called) from an external environment (e.g., facility 1302). In at least one embodiment, the applications may then call or execute one or more services 1320 to perform computational, AI, or visualization tasks associated with the respective application, and the software 1318 and / or services 1320 may utilize hardware 1322 to perform the processing tasks in an effective and efficient manner.

[0108] In at least one embodiment, the introduction system 1306 may execute an introduction pipeline 1410. In at least one embodiment, the introduction pipeline 1410 may include any number of applications, including the AI-assisted annotations described above, that may be applied sequentially, non-sequentially, or otherwise to imaging data (and / or other types of data) generated by an imaging device, a sequencing device, a genomics device, etc. In at least one embodiment, as described herein, the introduction pipeline 1410 for an individual device may be referred to as a virtual device for the device (e.g., a virtual ultrasound device, a virtual CT scan device, a virtual sequencing device, etc.). In at least one embodiment, there may be more than one introduction pipeline 1410 per device depending on the information required for the data generated by the device. In at least one embodiment, if anomaly detection is required for the MRI machine, a first introduction pipeline 1410 may be present, and if image enhancement is required for the output of the MRI machine, a second introduction pipeline 1410 may be present.

[0109] In at least one embodiment, the image generation application may include processing tasks that include the use of machine learning models. In at least one embodiment, a user may wish to use their own machine learning models or select a machine learning model from the model registry 1324. In at least one embodiment, a user may implement their own machine learning models or select a machine learning model to include in the application to perform the processing tasks. In at least one embodiment, the applications may be selectable and customizable, and by defining the structure of the application, the deployment and implementation of the application for a particular user may be presented as a more seamless user experience. In at least one embodiment, by utilizing other features of the system 1400, such as the services 1320 and the hardware 1322, the deployment pipeline 1410 may become even more user friendly, may provide easier integration, and may produce more accurate, efficient, and timely results.

[0110] In at least one embodiment, the deployment system 1306 may include a user interface 1414 (e.g., a graphical user interface, a web interface, etc.), which may be used to select applications for inclusion in the deployment pipeline 1410, configure applications, modify or change applications or their parameters or structure, use and interact with the deployment pipeline 1410 during setup and / or deployment, and / or otherwise interact with the deployment system 1306. In at least one embodiment, while not shown with respect to the training system 1304, the user interface 1414 (or a different user interface) may be used to select models for use in the deployment system 1306, to select models to train or retrain in the training system 1304, and / or to otherwise interact with the training system 1304.

[0111] In at least one embodiment, a pipeline manager 1412 may be used in addition to an application orchestration system 1428 to manage interactions between applications or containers in the deployment pipeline 1410 and services 1320 and / or hardware 1322. In at least one embodiment, the pipeline manager 1412 may be configured to facilitate application-to-application interactions, application-to-service interactions, and / or application or service-to-hardware interactions. While illustrated in at least one embodiment as being included in the software 1318, this is not intended to be limiting, and in some cases (e.g., such as those shown in FIG. 12cc), the pipeline manager 1412 may be included in the services 1320. In at least one embodiment, the application orchestration system 1428 (e.g., Kubernetes, DOCKER, etc.) may include a container orchestration system that may group applications into containers as logical units for orchestration, management, scaling, and deployment. In at least one embodiment, by associating applications from the deployment pipeline 1410 (e.g., a reconstruction application, a segmentation application, etc.) with individual containers, each application can run in a self-contained environment (e.g., at the kernel level) to improve speed and efficiency.

[0112] In at least one embodiment, each application and / or container (or an image thereof) may be developed, modified, and deployed individually (e.g., a first user or developer may develop, modify, and deploy a first application, and a second user or developer may develop, modify, and deploy a second application separately from the first user or developer), allowing focus and attention to be paid to the tasks of one application and / or container without being distracted by the tasks of another application or container. In at least one embodiment, communication and coordination between different containers or applications may be facilitated by the pipeline manager 1412 and the application orchestration system 1428. In at least one embodiment, the application orchestration system 1428 and / or the pipeline manager 1412 can facilitate communication between, and sharing of resources between, each of the applications or containers, so long as the expected inputs and / or outputs of each container or application are known by the system (e.g., based on the structure of the application or container). In at least one embodiment, one or more of the applications or containers in the deployment pipeline 1410 may share the same services and resources, so the application orchestration system 1428 may orchestrate, load balance, and determine sharing of services or resources among the various applications or containers. In at least one embodiment, a scheduler may be used to track the resource requirements of the applications or containers, the current or planned usage of those resources, and the availability of the resources. In at least one embodiment, the scheduler may thus allocate resources to different applications and distribute resources among the applications taking into account the requirements and availability of the system.In some instances, the scheduler (and / or other components of the application orchestration system 1428) may determine resource availability and distribution based on constraints imposed on the system (e.g., user constraints), such as quality of service (QoS), the urgency of needing data output (e.g., to determine whether to perform real-time or deferred processing), etc.

[0113] In at least one embodiment, services 1320 utilized and shared by applications or containers of deployment system 1306 may include compute services 1416, AI services 1418, visualization services 1420, and / or other types of services. In at least one embodiment, an application may call (e.g., execute) one or more of services 1320 to perform processing operations for the application. In at least one embodiment, compute services 1416 may be utilized by an application to perform supercomputing or other high performance computing (HPC) tasks. In at least one embodiment, parallel processing may be performed utilizing compute services 1416 (e.g., using parallel computing platform 1430) to substantially simultaneously process data via one or more of the applications and / or to substantially simultaneously process one or more tasks of an application. In at least one embodiment, parallel computing platform 1430 (e.g., NVIDIA's CUDA) may enable general purpose computing (GPGPU) on GPUs (e.g., GPU 1422). In at least one embodiment, the software layer of the parallel computing platform 1430 may provide access to virtual instruction sets and parallel computational elements of a GPU to execute computational kernels. In at least one embodiment, the parallel computing platform 1430 may include memory, and in some embodiments, memory may be shared among multiple containers and / or among different processing tasks within a container. In at least one embodiment, inter-process communication (IPC) calls may be generated in order for multiple containers and / or multiple processes within a container to use the same data from a shared segment of memory of the parallel computing platform 1430 (e.g., when multiple different stages of an application or multiple applications process the same information).In at least one embodiment, rather than making copies of the data and moving the data to a different location in memory (e.g., read / write operations), the same data in the same location in memory may be used for any number of processing tasks (e.g., at the same time, at different times, etc.). In at least one embodiment, as data is used and new data is generated as a result of processing, this information of the new location of the data may be stored in and shared between various applications. In at least one embodiment, the location of the data and the location of updated or modified data may be part of the definition of how the payload is understood within the container.

[0114] In at least one embodiment, the AI ​​services 1418 may be utilized to perform inference services to execute machine learning models associated with an application (e.g., tasked with performing one or more processing tasks of the application). In at least one embodiment, the AI ​​services 1418 may utilize the AI ​​systems 1424 to execute machine learning models (e.g., neural networks such as CNNs) for segmentation, reconstruction, object detection, feature detection, classification, and / or other inference tasks. In at least one embodiment, an application in the deployment pipeline 1410 may perform inference on imaging data using one or more of the output models 1316 from the training system 1304 and / or other models of the application. In at least one embodiment, two or more instances of inference using the application orchestration system 1428 (e.g., a scheduler) may be available. In at least one embodiment, a first category may include a high priority / low latency path that can achieve higher service level agreements, such as for performing inference on urgent requests during an emergency or for a radiologist during a diagnosis. In at least one embodiment, the second category may include a standard priority path that may be used for non-urgent requests or when the analysis may be performed at a later time. In at least one embodiment, the application orchestration system 1428 may allocate resources (e.g., services 1320 and / or hardware 1322) based on the priority paths for different inference tasks of the AI ​​service 1418.

[0115] In at least one embodiment, a shared storage may be attached to the AI ​​service 1418 in the system 1400. In at least one embodiment, the shared storage may act as a cache (or other type of storage device) and may be used to process inference requests from applications. In at least one embodiment, when an inference request is sent, the request may be received by a set of API instances of the installation system 1306, and one or more instances may be selected (e.g., for best fit, for load balancing, etc.) to process the request. In at least one embodiment, to process the request, the request may be entered into a database, a machine learning model may be identified from the model registry 1324 if not already in the cache, and a validation step may ensure that the appropriate machine learning model is loaded into the cache (e.g., the shared storage), and / or a copy of the model may be saved to the cache. In at least one embodiment, if the application is not already running or if there are not enough instances of the application, a scheduler (e.g., the pipeline manager 1412) may be used to launch the application referenced in the request. In at least one embodiment, an inference server may be started to run the model if one has not already been started. Any number of inference servers may be started per model. In at least one embodiment, in a pull model where inference servers are clustered, models may be cached whenever load balancing is advantageous. In at least one embodiment, inference servers may be statically loaded onto corresponding distributed servers.

[0116] In at least one embodiment, inference may be performed using an inference server running within a container. In at least one embodiment, an instance of an inference server may be associated with a model (and optionally with multiple versions of the model). In at least one embodiment, when a request to perform inference on a model is received, if an instance of the inference server does not exist, a new instance may be loaded. In at least one embodiment, when the inference server is started, the model may be passed to the inference server, such that the same container may be used to serve different models, as long as the inference servers are running as different instances.

[0117] In at least one embodiment, while an application is running, an inference request may be received for a given application, a container (e.g., hosting an instance of an inference server) may be loaded (if not already loaded) and a start procedure may be called. In at least one embodiment, pre-processing logic of the container may load, decode, and / or perform any additional pre-processing on the input data (e.g., using a CPU and / or GPU). In at least one embodiment, once the data is prepared for inference, the container may perform inference on the data as needed. In at least one embodiment, this may include a single inference call for one image (e.g., a hand x-ray) or may require inference on hundreds of images (e.g., a chest CT). In at least one embodiment, the application may summarize the results before completion, which may include, without limitation, a single confidence score, pixel-level segmentation, voxel-level segmentation, generation of a visualization, or generation of text to summarize the findings. In at least one embodiment, different models or applications may be assigned different priorities. For example, some models may have real-time priority (TAT<1 minute) and other models may have low priority (e.g., TAT<10 minutes). In at least one embodiment, model execution time may be measured from the requesting facility or entity and may include execution to the inference service as well as partner network traversal time.

[0118] In at least one embodiment, the transition of requests between the service 1320 and the inference application may be hidden behind a software development kit (SDK) and robust transport may be provided through queues. In at least one embodiment, requests are queued through an API for individual application / tenant ID combinations and the SDK pulls requests from the queue and gives them to the application. In at least one embodiment, a name for the queue may be provided in the environment where the SDK picks up the request. In at least one embodiment, asynchronous communication through the queue may be useful as it allows any instance of the application to pick up work when it is available. Results may be sent back through the queue to ensure data is not lost. In at least one embodiment, queues may also provide the ability to segment work, as the highest priority work may go to a queue with the most instances of the application connected to the queue, while the lowest priority work may go to a queue with one instance connected to the queue that processes tasks in the order they are received. In at least one embodiment, applications may run on GPU accelerated instances spawned in the cloud 1426 and the inference service may run inference on the GPU.

[0119] In at least one embodiment, visualization services 1420 may be utilized to generate visualizations for viewing applications and / or the output of deployment pipeline 1410. In at least one embodiment, GPU 1422 may be utilized by visualization services 1420 to generate visualizations. In at least one embodiment, rendering effects such as ray tracing may be implemented by visualization services 1420 to generate higher quality visualizations. In at least one embodiment, visualizations may include, without limitation, 2D image rendering, 3D volume rendering, 3D volume reconstruction, 2D tomography slices, virtual reality displays, augmented reality displays, and the like. In at least one embodiment, virtualized environments may be used to generate virtual interactive displays or environments (e.g., virtual environments) for users of the system to interact with. In at least one embodiment, visualization services 1420 may include internal visualizers, cinematics, and / or other rendering or image processing capabilities or functions (e.g., ray tracing, rasterization, internal optics, etc.).

[0120] In at least one embodiment, hardware 1322 may include GPU 1422, AI system 1424, cloud 1426, and / or any other hardware used to run training system 1304 and / or deployment system 1306. In at least one embodiment, GPU 1422 (e.g., NVIDIA TESLA and / or QUADRO GPUs) may include any number of GPUs that may be used to perform processing tasks for compute services 1416, AI services 1418, visualization services 1420, other services, and / or any features or functionality of software 1318. For example, with respect to AI services 1418, GPU 1422 may be used to perform pre-processing on imaging data (or other types of data used by machine learning models), post-processing on the output of machine learning models, and / or inference (e.g., machine learning models may be executed). In at least one embodiment, cloud 1426, AI system 1424, and / or other components of system 1400 may use GPU 1422. In at least one embodiment, cloud 1426 may include a GPU-optimized platform for deep learning tasks. In at least one embodiment, AI system 1424 may use GPUs, and cloud 1426, or at least a portion tasked with deep learning or inference, may be executed using one or more AI systems 1424. Thus, while hardware 1322 is shown as separate components, this is not intended to be limiting, and any component of hardware 1322 may be combined with or utilized by any other component of hardware 1322.

[0121] In at least one embodiment, the AI ​​system 1424 may include a specialized computing system (e.g., a supercomputer or HPC) configured for inference, deep learning, machine learning, and / or other artificial intelligence tasks. In at least one embodiment, the AI ​​system 1424 (e.g., NVIDIA's DGX) may include GPU-optimized software (e.g., a software stack) that may be executed using multiple GPUs 1422 in addition to CPUs, RAM, storage, and / or other components, features, or functions. In at least one embodiment, one or more AI systems 1424 may be implemented in the cloud 1426 (e.g., in a data center) to perform some or all of the AI-based processing tasks of the system 1400.

[0122] In at least one embodiment, the cloud 1426 may include a GPU-accelerated infrastructure (e.g., NVIDIA's NGC), which may provide a GPU-optimized platform for executing the processing tasks of the system 1400. In at least one embodiment, the cloud 1426 may include an AI system 1424 (e.g., as a hardware abstraction and scaling platform) for executing one or more of the AI-based tasks of the system 1400. In at least one embodiment, the cloud 1426 may utilize multiple GPUs and integrate with an application orchestration system 1428 to enable seamless scaling and load balancing between applications and services 1320. In at least one embodiment, the cloud 1426 may be tasked with executing at least a portion of the services 1320 of the system 1400, including the compute services 1416, the AI ​​services 1418, and / or the visualization services 1420 described herein. In at least one embodiment, cloud 1426 may perform small or large batch inference (e.g., NVIDIA's Tensor RT implementation), may provide an accelerated parallel computing API and platform 1430 (e.g., NVIDIA's CUDA), may run an application orchestration system 1428 (e.g., KUBERNETES), may provide a graphics rendering API and platform (e.g., ray tracing, 2D graphics, 3D graphics, and / or other rendering techniques for generating high quality cinematics), and / or may provide other functionality for system 1400.

[0123] 15A illustrates a data flow diagram of a process 1500 for training, retraining, or updating a machine learning model, according to at least one embodiment. In at least one embodiment, the process 1500 may be performed using the system 1400 of FIG. 14 as a non-limiting example. In at least one embodiment, the process 1500 may utilize the services 1320 and / or hardware 1322 of the system 1400 described herein. In at least one embodiment, the refined model 1512 generated by the process 1500 may be executed by the deployment system 1306 for one or more containerized applications in the deployment pipeline 1410.

[0124] In at least one embodiment, model training 1314 may include retraining or updating the initial model 1504 (e.g., a pre-trained model) using new training data (e.g., new input data, such as the customer dataset 1506 and / or new ground truth data associated with the input data). In at least one embodiment, to retrain or update the initial model 1504, an output or loss layer of the initial model 1504 may be reset, removed, and / or replaced with an updated or new output or loss layer. In at least one embodiment, the initial model 1504 may have previously fine-tuned parameters (e.g., weights and / or biases) remaining from a previous training, such that training or retraining 1314 does not take as long or require as much processing as training a model from scratch. In at least one embodiment, during model training 1314, by having the output or loss layer of the initial model 1504 reset or replaced, parameters may be updated or retuned for the new data set based on a loss calculation associated with the accuracy of the output or loss layer in generating predictions for the new customer data set 1506 (e.g., image data 1308 in FIG. 13 ).

[0125] In at least one embodiment, the pre-trained model 1406 may be stored in a data store or registry (e.g., model registry 1324 of FIG. 13). In at least one embodiment, the pre-trained model 1406 may be trained, at least in part, at one or more facilities different from the facility performing the process 1500. In at least one embodiment, to protect the privacy and rights of patients, subjects, or customers of the different facilities, the pre-trained model 1406 may have been trained on-premise using customer or patient data generated on-premise. In at least one embodiment, the pre-trained model 1406 may be trained using the cloud 1426 and / or other hardware 1322, but the privacy-protected sensitive patient data may not be transferred to, used by, or accessible by any components of the cloud 1426 (or other off-premise hardware). In at least one embodiment, if the pre-trained model 1406 is trained using patient data from more than one facility, the pre-trained model 1406 may be trained individually for each facility and then trained on patient or customer data from another facility. In at least one embodiment, the pre-trained model 1406 may be trained on-premise and / or off-premise, such as in a data center or other cloud computing infrastructure, using customer or patient data from any number of facilities, such as if the customer or patient data is free of privacy concerns (e.g., via a waiver for experimental use) or if the customer or patient data is included in a public dataset.

[0126] In at least one embodiment, when selecting an application for use in the deployment pipeline 1410, the user may also select the machine learning model that will be used with the particular application. In at least one embodiment, the user may not have a model to use, and therefore the user may select a pre-trained model 1406 to use with the application. In at least one embodiment, the trained model 1406 may be optimized to produce accurate results for the user's facility's customer dataset 1506 (e.g., based on patient diversity, demographics, type of medical imaging device used, etc.). In at least one embodiment, before the pre-trained model 1406 is deployed in the deployment pipeline 1410 for use with the application, the pre-trained model 1406 may be updated, retrained, and / or fine-tuned for use at the respective facility.

[0127] In at least one embodiment, a user may select a pre-trained model 1406 to be updated, retrained, and / or fine-tuned, which may be referred to as an initial model 1504 for training the system 1304 in the process 1500. In at least one embodiment, model training 1314 (which may include, without limitation, transfer learning) may be performed on the initial model 1504 using a customer dataset 1506 (e.g., imaging data, genomics data, sequencing data, or other types of data generated by devices at the facility) to generate a refined model 1512. In at least one embodiment, ground truth data corresponding to the customer dataset 1506 may be generated by the training system 1304. In at least one embodiment, the ground truth data may be generated, at least in part, by clinicians, scientists, physicians, practitioners at the facility (e.g., as labeled clinic data 1312 in FIG. 13 ).

[0128] In at least one embodiment, AI-assisted annotation 1310 may be used in some instances to generate ground truth data. In at least one embodiment, AI-assisted annotation 1310 (e.g., implemented using an AI-assisted annotation SDK) may utilize machine learning models (e.g., neural networks) to generate suggested or predicted ground truth data for a customer dataset. In at least one embodiment, a user 1510 may use an annotation tool within a user interface (graphical user interface (GUI)) on a computing device 1508.

[0129] In at least one embodiment, a user 1510 may interact with a GUI via a computing device 1508 to edit or fine-tune the (automatic) annotation. In at least one embodiment, polygon editing features may be used to move vertices of polygons to more accurate or fine-tuned locations.

[0130] In at least one embodiment, once the customer dataset 1506 has associated ground truth data, the ground truth data (e.g., from AI-assisted annotation, manual labeling, etc.) may be used during model training 1314 to generate a refined model 1512. In at least one embodiment, the customer dataset 1506 may be applied to the initial model 1504 any number of times, and the ground truth data may be used to update parameters of the initial model 1504 until an acceptable level of accuracy is achieved for the refined model 1512. In at least one embodiment, once the refined model 1512 is generated, the refined model 1512 may be deployed into one or more deployment pipelines 1410 at the facility to perform one or more processing tasks on the medical imaging data.

[0131] In at least one embodiment, the refined model 1512 may be uploaded to the pre-trained models 1406 of the model registry 1324 to be selected by another facility. In at least one embodiment, this process may be completed at any number of facilities, such that the refined model 1512 may be further refined any number of times on new datasets to generate a more general model.

[0132] FIG. 15B is a diagram of an example client-server architecture 1532 for enhancing an annotation tool with a pre-trained annotation model, according to at least one embodiment. In at least one embodiment, an AI-assisted annotation tool 1536 may be instantiated based on the client-server architecture 1532. In at least one embodiment, the annotation tool 1536 of an imaging application may assist, for example, a radiologist in identifying organs and anomalies. In at least one embodiment, the imaging application may include, as a non-limiting example, a software tool that assists a user 1510 in identifying a few extreme points on a particular target organ in a raw image 1534 (e.g., of a 3D MRI or CR scan) and receives an automatically annotated result for all 2D slices of the particular organ. In at least one embodiment, the results may be stored in a data store as training data 1538 or may be used (for example, without limitation) as ground truth data for training. In at least one embodiment, when the computing device 1508 sends the extreme points for AI-assisted annotation 1310, for example, a deep learning model may receive this data as input and return an inference result of a segmented organ or anomaly. In at least one embodiment, a pre-instantiated annotation tool, such as the AI-assisted annotation tool 1536B of FIG. 15B, may be extended by making an API call (e.g., API call 1544) to a server, such as an annotation assisted server 1540, which may include a set of pre-trained models 1542 stored in an annotation model registry. In at least one embodiment, the annotation model registry may store pre-trained models 1542 (e.g., machine learning models such as deep learning models) that have been pre-trained to perform AI-assisted annotation for a particular organ or anomaly. These models may be further updated by using the training pipeline 1404.In at least one embodiment, the pre-installed annotation tools may be improved over time as new labeled clinic data 1312 is added.

[0133] Using these components, live updates of machine learning models can be performed for applications running at locations such as the network edge, without any downtime as the application switches to a different model version, or without any significant downtime.

[0134] Other variations are within the scope of the present disclosure. Thus, while the disclosed techniques are susceptible to various modifications and alternative constructions, certain illustrative examples thereof have been shown in the drawings and have been described in detail above. However, there is no intention to limit the disclosure to the particular disclosed form or forms, but on the contrary, the intention is to cover all modifications, alternative constructions, and equivalents falling within the spirit and scope of the disclosure as defined by the appended claims.

[0135] The use of the terms "a" and "an" and "the" and similar referents in the context of describing the disclosed embodiments (particularly in the context of the claims that follow) should be construed to cover both the singular and the plural and should not be construed as defining terms, unless otherwise stated herein or clearly contradicted by context. The terms "comprising," "having," "including," and "containing" are to be construed as open-ended terms (meaning "including, but not limited to"), unless otherwise stated. The term "connected," when referring to a physical connection without modification, is to be construed as being partially or completely contained within, attached to, or joined to one another, even if there is something intervening. The recitation of ranges of values ​​herein is merely intended to serve as a shorthand method of individually referring to each separate value falling within the range, unless otherwise stated herein and unless each separate value is incorporated into the specification as if it were individually recited herein. Use of the term "set" (e.g., "set of items") or "subset" should be construed as a non-empty collection having one or more members, unless otherwise indicated or constrained by context. Further, unless otherwise indicated or constrained by context, the term "subset" of a corresponding set does not necessarily refer to a strict subset of the corresponding set, and a subset and a corresponding set may be equivalent.

[0136] Conjunctions such as phrases of the form "at least one of A, B, and C" or "at least one of A, B, and C" are understood in the context in which they are generally used to indicate that an item, term, etc. is A, B, or C, or a non-empty subset of any of the sets A, B, and C, unless specifically stated otherwise or clearly contradicted by context. For example, in the illustrative example of a set having three members, the conjunctive phrases "at least one of A, B, and C" and "at least one of A, B, and C" refer to any of the following sets: {A}, {B}, {C}, {A, B}, {A, C}, {B, C}, {A, B, C}. Thus, such conjunctives do not generally imply that a given embodiment requires the presence of at least one A, at least one B, and at least one C, respectively. Further, unless specifically stated otherwise or clearly stated by context, the term "plurality" refers to a plurality (e.g., "a plurality of items" refers to multiple items). A plurality is at least two items, but may be more if indicated explicitly or by context. Further, unless specifically stated otherwise or clearly stated otherwise by context, the phrase "based on" means "based at least in part on" and not "based only on."

[0137] The operations of the processes described herein may be performed in any suitable order unless otherwise stated herein or clearly contradicted by context. In at least one embodiment, a process such as the processes described herein (or variations and / or combinations thereof) is implemented as code (e.g., executable instructions, one or more computer programs, or one or more applications) executed under the control of one or more computer systems configured with executable instructions and collectively executed on one or more processors, by hardware, or by a combination thereof. In at least one embodiment, the code is stored on a computer-readable storage medium, e.g., in the form of a computer program comprising a plurality of instructions executable by one or more processors. In at least one embodiment, the computer-readable storage medium is a non-transitory computer-readable storage medium that excludes transitory signals (e.g., propagating transitory electrical or electromagnetic transmissions), but includes non-transitory data storage circuitry (e.g., buffers, caches, and queues) within a transceiver of a transitory signal. In at least one embodiment, code (e.g., executable code or source code) is stored on a set of one or more non-transitory computer readable storage media that store executable instructions (or have other memory for storing executable instructions) that, when executed by (i.e., as a result of being executed by) one or more processors of a computer system, cause the computer system to perform the operations described herein. The set of non-transitory computer readable storage media, in at least one embodiment, comprises a plurality of non-transitory computer readable storage media, where one or more of the individual non-transitory storage media of the plurality of non-transitory computer readable storage media does not contain all of the code, but the plurality of non-transitory computer readable storage media collectively store all of the code.In at least one embodiment, the executable instructions are executed such that different instructions are executed by different processors, e.g., a non-transitory computer-readable storage medium stores the instructions, a main central processing unit ("CPU") executes some instructions, and a graphics processing unit ("GPU") executes other instructions. In at least one embodiment, different components of a computer system have separate processors, and the different processors execute different subsets of the instructions.

[0138] Thus, in at least one embodiment, a computer system is configured to implement one or more services that, singly or collectively, perform the operations of the processes described herein, and such computer system is configured with applicable hardware and / or software that enables the operations to be performed. Further, a computer system that implements at least one embodiment of the present disclosure may be a single device, or in another embodiment, a distributed computer system comprising multiple devices that operate in different ways, such that the distributed computer system performs the operations described herein such that no single device performs all the operations.

[0139] Any examples provided herein, or the use of exemplary language (e.g., "such as"), are intended only to further clarify embodiments of the disclosure and do not limit the scope of the disclosure unless otherwise stated. No language in the specification should be construed as indicating any non-claimed element as essential to the practice of the disclosure.

[0140] All references cited in this specification, including publications, patent applications, and patents, are hereby incorporated by reference to the same extent as if each reference was individually indicated to be incorporated by reference and was set forth in its entirety herein.

[0141] In the specification and claims, the terms "coupled" and "connected," along with their derivatives, may be used. It should be understood that these terms may not be intended as synonyms for each other. Rather, in certain instances, "connected" or "coupled" may be used to indicate that two or more elements are in direct or indirect physical or electrical contact with each other. "Coupled" may also mean that two or more elements are not in direct contact with each other, but still interlock or interact with each other.

[0142] Unless specifically noted otherwise, terms such as "processing," "computing," "calculating," or "determining" throughout the specification refer to the acts and / or processes of a computer or computing system, or similar electronic computing device, that manipulate and / or transform data represented as physical quantities, such as electronic quantities, in the registers and / or memory of the computing system into other data similarly represented as physical quantities in the memory, registers, or other such information storage, transmission, or display device of the computing system.

[0143] Similarly, the term "processor" may refer to any device, or part of a device, that processes electronic data from registers and / or memory and converts the electronic data into other electronic data that can be stored in registers and / or memory. As a non-limiting example, a "processor" may be a CPU or a GPU. A "computing platform" may comprise one or more processors. A "software" process as used herein may include software and / or hardware entities that perform work over time, such as tasks, threads, and intelligent agents. Also, each process may refer to multiple processes for executing instructions serially or in parallel, continuously or intermittently. The terms "system" and "method" are used interchangeably herein to the extent that a system may embody one or more methods and a method may be considered a system.

[0144] As used herein, reference may be made to obtaining, acquiring, receiving, or inputting analog or digital data into a subsystem, computer system, or computer-implemented machine. Obtaining, acquiring, receiving, or inputting analog or digital data may be accomplished in various ways, such as receiving the data as a parameter of a function call or a call to an application programming interface. In some implementations, the process of obtaining, acquiring, receiving, or inputting analog or digital data may be accomplished by transferring the data over a serial or parallel interface. In other implementations, the process of obtaining, acquiring, receiving, or inputting analog or digital data may be accomplished by transferring the data over a computer network from a providing entity to an acquiring entity. Reference may also be made to providing, outputting, transmitting, sending, or presenting analog or digital data. In various examples, the process of providing, outputting, transmitting, sending, or presenting analog or digital data may be accomplished by transferring the data as an input or output parameter of a function call, a parameter of an application programming interface, or an inter-process communication mechanism.

[0145] Although the above discussion describes exemplary implementations of the described techniques, other architectures may be used to implement the described functionality, and these other architectures are intended to be within the scope of the present disclosure. Moreover, while a specific distribution of roles has been defined for purposes of discussion, the various functions and roles may be distributed and divided in different manners, depending on the circumstances. Moreover, while subject matter has been described in language specific to structural features and / or methodological acts, it should be understood that the subject matter claimed in the appended claims is not necessarily limited to the specific features or acts described. Rather, the specific features and acts are disclosed as exemplary forms of implementing the claims.

Claims

1. running an application on an edge computing device using a first version of the machine learning model; receiving a second version of the machine learning model at the edge computing device; receiving new configuration data for the application at the edge computing device that specifies use of the second version; In response to detecting the new configuration data, causing the application to automatically switch to the second version of the machine learning model while running, without stopping the application.

4. A computer-implemented method comprising:

2. 2. The computer-implemented method of claim 1, wherein the edge computing device is a server or a system-on-chip (SoC) located at a network edge.

3. 2. The computer-implemented method of claim 1, wherein the first version and the second version are stored in a model directory of a storage volume mounted on the edge computing device.

4. monitoring a model store associated with the machine learning model; automatically fetching the second version of the machine learning model to the edge computing device in response to detecting the second version in the model store; The computer-implemented method of claim 1 , further comprising:

5. 2. The computer-implemented method of claim 1, further comprising the step of generating a new context in response to detecting the new configuration data, further responsive to the application detecting that the new context is available, causing the application to automatically switch to using the second version.

6. The computer-implemented method of claim 1 , further comprising enabling the application to automatically switch to using the second version without requiring a restart of the application.

7. 2. The computer-implemented method of claim 1 , wherein the application was deployed to the edge computing device as part of an application container that did not include the first version of the machine learning model.

8. mounting the second version of the machine learning model from local storage on the edge computing device to an application container in which the application is running, wherein the application can automatically switch to the second version of the machine learning model; removing the first version of the machine learning model from the application container but retaining the first version in the local storage of the edge computing device; The computer-implemented method of claim 7 , further comprising:

9. receiving updated configuration data for the application that specifies use of the first version; In response to detecting the updated configuration data, causing the application to automatically switch to the first version of the machine learning model during runtime, wherein the first version is mounted from the local storage of the edge computing device to the application container; The computer-implemented method of claim 8 , further comprising:

10. 1. A system comprising: At least one processor; When executed by the at least one processor, the system comprises: running an application using a first version of the machine learning model; receiving a second version of the machine learning model; receiving new configuration data for the application that specifies use of the second version; and In response to detecting the new configuration data, causing the application to automatically switch to the second version of the machine learning model during execution without stopping the application, where the second version of the machine learning model is allowed to be mounted from local storage into an application container in which the application is executing. a memory containing instructions for performing A system including:

11. The instructions, when executed, further cause the system to: monitoring a model store associated with the machine learning model; automatically fetching the second version of the machine learning model to a computing device in response to detecting the second version in the model store; The system of claim 10 .

12. the system is located at a network edge; 11. The system of claim 10, further comprising: local storage for storing the first version and the second version in a subdirectory of a model directory, the local storage, when mounted in the application container, enabling the application to automatically switch to the second version of the machine learning model.

13. 13. The system of claim 12, wherein the application was deployed to the system as part of an application container that did not include the first version of the machine learning model.

14. The instructions, when executed, further cause the system to:

11. The system of claim 10, further comprising: generating a new context in response to detecting the new configuration data; and further causing the application to automatically switch to using the second version in response to the application detecting that the new context is available.

15. The system of claim 10 , wherein the application is enabled to automatically switch to using the second version without requiring a restart of the application.

16. running an application in an application container on a network edge device utilizing a first version of the machine learning model for inference on the data stream; detecting a new version of the machine learning model available from a model source; fetching the new version of the model to a local storage of the network edge device; updating context information to indicate the new version of the model; mounting the new version of the model into the application container; In response to updating the context information, automatically switching the application to utilize the new version of the machine learning model without losing data from the data stream and without stopping the application while it is running.

23. A computer-implemented method comprising:

17. 17. The computer-implemented method of claim 16, wherein the first version and the new version are stored in subdirectories of a models directory of a storage volume mounted on the network edge device.

18. 17. The computer-implemented method of claim 16, further comprising causing the application to detect the updated context information and, in response, automatically switch to utilize the new version without restarting or updating the application.

19. 20. The computer-implemented method of claim 16, wherein the application was deployed to the network edge device as part of an application container that did not include the first version of the machine learning model.

20. 20. The computer-implemented method of claim 16, further comprising removing the first version of the machine learning model from the application container but retaining the first version in the local storage of the computing device.

Images