Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

20090results about "Physical realisation" patented technology

Convergent Intelligence Fabric for Multi-Domain Orchestration of Distributed Agents with Hierarchical Memory Architecture and Quantum-Resistant Trust Mechanisms

A system and method for implementing a convergent intelligence fabric (CIF) for distributed artificial intelligence operations. The CIF architecture integrates tensor-theoretic foundations, probabilistic cache management, precision-aware memory operations, quantum-resistant security, and neural-based optimization within a unified framework. The system orchestrates asynchronous, multi-hop data flow among computational resources while maintaining data security through per-block encryption and identity-based access control. Key components include a universal multi-model KV cache subsystem, agent-parallel disaggregation pipelines, reinforcement learning-based orchestration, and neuromorphic memory integration. Advanced implementations incorporate graphon-enhanced memory for sparse graph sequences, multi-modal cognitive persistent memory, and quantum-resistant asynchronous multi-domain trust protocols. The system enables efficient cross-agent collaboration, sophisticated knowledge sharing, and secure cross-domain operations while optimizing computational resources and maintaining strict privacy guarantees across distributed AI deployments.
Owner:QOMPLX INC

AI Serving Hardware and Software Frontier Enhancements

A computer system implements a unified framework integrating an adaptive elastic funnel (AEF) with a convergent intelligence fabric (CIF) for multi-agent AI collaboration. The system provides a universal multi-modal key-value subsystem for sharing partial computations, implements hybrid placement strategies for dynamic memory management, and incorporates quantum-resistant secure enclaves. The architecture integrates hardware acceleration through GPU-FPGA hybrid caching and neuromorphic processors, applies adaptive energy and thermal management across hardware generations, and implements autonomous flash resource orchestration with multi-dimensional wear management. The system orchestrates tensor workflows using hierarchical scheduling, enables cross-agent collaboration with privacy preservation, and supports continuous learning without catastrophic forgetting. This integration delivers unprecedented computational efficiency and security in high-dimensional decision-making environments while supporting incremental adoption through modular interfaces.
Owner:QOMPLX INC

Systems and Methods for Dynamic Neural Network Enhancement and Adaptive Edge Computing

Systems and methods for adaptive edge computing using artificial intelligence (AI) include monitoring real-time accuracy of a neural network by using a feedback loop configured to detect changes in inference accuracy and dynamically adjusting the structure of the neural network by adding or removing hidden layers based on monitored error rates and predetermined computational constraints. A Kalman gain computation determines neural network weight adjustments based on monitored error rates. Weight matrices undergo incremental updates derived from these adjustments. Incremental weight adjustments remain stored in memory to enable low-bandwidth model updates. The neural network stores inference results and refined weights in an inference result database. Pre-trained models periodically receive incremental updates based on stored adjustments. Predictive holistic inference logic (PHIL) applied to stored inference results improves the accuracy of the inference results.
Owner:VEEA INC

Collaborative task offloading and service caching method based on graph attention multi-agent reinforcement learning

PCT designated stage expiredWO2025050608A1Mathematical modelsNetwork traffic/resource management
A collaborative task offloading and service caching method based on graph attention multi-agent reinforcement learning, which method belongs to the technical field of mobile communications. The method comprises: constructing an MEC system in an MEC network scenario where there are multiple base stations and diversified service requests; modeling a collaborative task offloading and service caching problem by taking as an objective the maximization of a system utility based on QoE; converting the collaborative task offloading and service caching problem into a decentralized partially observable Markov decision process; and using an algorithm based on graph attention multi-agent reinforcement learning to solve the decentralized partially observable Markov decision process obtained by means of conversion. By means of a simulation experiment, compared with a benchmark algorithm, significant improvements are made in terms of a cache hit rate, system utility and computation success rate to the algorithm proposed herein.
Owner:CHONGQING UNIV OF POSTS & TELECOMM

Dynamic quantization and memory management of key-value cache for serving large language models

Key-value (KV) cache paging schemes can improve memory management for KV caches by storing a KV cache page having key tensors and value tensors for a fixed number of tokens in a fixed-sized block in the KV cache of a worker. To further improve memory management, the schemes can be modified to implement dynamic variable quantization. Quantization level of a KV cache page can be set based on a runtime importance score of the KV cache page. In addition, the quantization level of the KV cache page can be set based on the system load. The end result is a scheme that can achieve a high compression ratio of KV cache pages in the KV cache. Fitting more KV cache pages in the KV cache can lead to higher inference throughput, higher system-level user capacity, and higher end-to-end service availability.
Owner:INTEL CORP

Systems, methods, kits, and apparatuses for specialized chips for robotic intelligence layers

A system may include a robotic control circuit configured to control one or more robotic functions of a robot. A system may include a plurality of sensors configured to collect data. A system may include a governance analysis circuit configured to analyze the data and select one or more governance frameworks based on the analyzed data. A system may include a governance model circuit configured to generate a model that applies the one or more governance frameworks to determine one or more governance actions, wherein the robotic control circuit is configured to control the one or more robotic functions in accordance with the one or more governance actions, wherein the robotic control circuit, the governance analysis circuit, and the governance model circuit are integrated on a single substrate.
Owner:STRONG FORCE VCN PORTFOLIO 2019 LLC

Load balancing method of low-power AI processor, chip and storage medium

The invention relates to the technical field of low-power-consumption AI processors, and discloses a load balancing method of a low-power-consumption AI processor, a chip and a storage medium, the method comprises the following steps: performing cache resource division on a plurality of neural mimicry cores in the low-power-consumption AI processor to obtain a double-layer cache structure; performing characteristic analysis on the target calculation task, and constructing a calculation task directed acyclic graph and a dependency intensity matrix; performing task segmentation to obtain a subtask division scheme; creating a task-core mapping table; organizing the neural mimicry core into a plurality of concentric ring layers based on the task-core mapping table to obtain a multi-layer ring cooperative computing network; and calculating a plurality of performance indexes of each neural mimicry core based on the multilayer annular cooperative computing network, and performing dynamic load optimization to obtain a dynamic load optimization strategy. According to the invention, the load balancing of the low-power AI processor is realized, and the calculation efficiency and the energy utilization rate are improved.
Owner:SHENZHEN ITZR TECH

Converting quasi-affine expressions to matrix operations

A computer-implemented method includes generating, based on a representation of a tensor mapping between an input tensor and an output tensor, a list of mappings from elements of the input tensor to elements of the output tensor, and generating groups of mappings from the list of mappings, where each of the groups of mappings corresponds to a respective set of matrix multiplications, a matrix transpose, or both. The computer-implemented method also includes generating a respective expression for each of the groups of mappings and generating code for summing results of the respective expressions, where each respective expression includes the respective set of matrix multiplications, the matrix transpose, or both.
Owner:AMAZON TECH INC

Large model reasoning efficiency dynamic optimization and hardware sensing compression method

The invention discloses a large model reasoning efficiency dynamic optimization and hardware sensing compression method. The method comprises the following five steps: S1, generating an input complexity signal representing calculation complexity; s2, synchronously monitoring a hardware resource index of the operation platform, and generating a hardware state signal reflecting a real-time load; s3, inputting the input complexity signal and the hardware state signal into a dynamic strategy selector, and generating a compression control signal through a pre-trained decision model; s4, according to the compression control signal, dynamic reconfiguration operation is executed on the large model weight and the activation value of the current reasoning task; and S5, performing reasoning calculation by using the reconfigured large model, and feeding back a hardware resource index to the step S2 in real time in the calculation process to form a closed-loop optimization link. According to the large model reasoning efficiency dynamic optimization and hardware perception compression method, the problems of low resource utilization rate, delay fluctuation and energy efficiency imbalance caused by a static compression method in dynamic input and heterogeneous hardware environments can be solved.
Owner:KARAMAY HONGYOU SOFTWARE

Parallel strategy search method for efficient training of artificial intelligence large model

The invention discloses a parallel strategy search method for efficient training of an artificial intelligence large model, and relates to the technical field of artificial intelligence, and the method comprises the steps: generating a three-dimensional hardware topology field through training a NeRF model; analyzing the calculation graph of the large model to be trained, mapping operator nodes into the three-dimensional hardware topology field, and generating a labeling calculation graph; constructing a mixed integer programming model based on the annotation calculation graph, and outputting a candidate strategy set; inputting the candidate strategy set into a double-layer strategy network, and outputting a parallel strategy set; loading the three-dimensional hardware topology field in the simulator, verifying and adjusting the parallel strategy set, and outputting an optimal strategy; and deploying the optimal strategy to the real cluster, and triggering a re-screening strategy. According to the method, a physical-level hardware communication field is modeled based on a neural radiation field, mixed integer programming and deep reinforcement learning are fused, parallel strategy global-local collaborative optimization is realized, and multi-target balance of training efficiency, video memory utilization rate and system robustness is realized.
Owner:西安圣瞳科技有限公司

Big data-based energy storage power station battery health state diagnosis method and system

The invention provides an energy storage power station battery health state diagnosis method and system based on big data, and relates to the technical field of energy storage batteries, and the method comprises the steps: collecting multi-mode operation data of an energy storage power station battery, carrying out homomorphic encryption to obtain ciphertext data, carrying out analysis and calculation, extracting explicit and implicit correlation features, and obtaining fusion time sequence data; based on fusion time sequence data, extracting multi-path quantum entanglement features, combining with a self-evolution quantum attention mechanism, dynamically adjusting the phase and amplitude of quantum bits, generating diagnosis feature enhancement representation, realizing entanglement storage through a quantum circuit, and combining with a gating unit to predict a health state sequence; and determining a battery health prediction result, and obtaining a battery health state diagnosis result by combining quantum state superposition and quantum parallelism, searching a diagnosis strategy combination space on line, embedding a multi-order causal relationship tensor model, modeling causal relationship strength between battery performance indexes, determining a battery health degradation index, generating a causal diagnosis map and obtaining the battery health state diagnosis result.
Owner:JIANGSU ZHONGJUN ENERGY TECH CO LTD

Neural network based physical condition evaluation of electronic devices, and associated systems and methods

Systems and methods for evaluating the physical and / or cosmetic condition of electronic devices using machine learning techniques are disclosed. In one example aspect, an example system includes a kiosk that comprises an inspection plate configured to hold an electronic device, one or more light sources arranged above the inspection plate configured to direct one or more light beams towards the electronic device, and one or more cameras configured to capture at least one image of a first side of the electronic device. The system also includes one or more processors in communication with the one or more cameras configured to extract a set of features of the electronic device and determine, via a first neural network, a condition of the electronic device based on the set of features.
Owner:ECOATM LLC

Using vector clocks to simplify a dependency graph in a neural network accelerator

Methods for simplifying a dependency graph in a neural network accelerator are provided. Computations and data movements for the neural network accelerator may be described with a flow graph, where graph nodes represent computation or data movement operations and graph edges represent dependencies between operations. A flow graph may contain redundant edges that can be removed while retaining the reachability of each of the nodes in the graph. To identify redundant edges, a compiler may generate vector clocks to track the relationships of operations performed by various execution engines prior to execution of a program reaching a given node or operation. Redundant edges may be identified and removed based on the relative values of the vector clocks to reduce the complexity of the graph.
Owner:AMAZON TECH INC

Heterogeneous computing acceleration method and system based on deep learning framework network

The invention relates to the technical field of data processing, and discloses a heterogeneous computing acceleration method and system based on a deep learning framework network. The method comprises the following steps: acquiring performance parameters of heterogeneous computing equipment to obtain an equipment characteristic data set; receiving a calculation task and analyzing the calculation task into an operation sequence; the operation sequence is coded into a gene sequence, task decomposition is optimized through a genetic recombination algorithm, and a subtask set marked with acceleration characteristics is obtained; performing matching analysis on the sub-task set and the equipment characteristic data set to obtain a task allocation scheme; deploying the subtasks to corresponding equipment according to the allocation scheme to obtain a distributed execution framework; and monitoring the running state of the framework in real time, and dynamically adjusting resource allocation to obtain a calculation result of speed-up ratio improvement. According to the method, a self-adaptive task decomposition and resource allocation and dynamic load balancing mechanism can be realized, and the execution efficiency and the resource utilization rate of the deep learning task are improved.
Owner:无锡九方科技有限公司

Magnet surface defect detection equipment control method for AI optimization and controller

The invention discloses a magnet surface defect detection equipment control method for AI optimization and a controller, and relates to the field of control adjusting systems.The method comprises the steps that according to a conveying monitoring image, surface brightness information of a to-be-detected magnet on a conveying platform is determined; determining detection light source parameters according to the surface brightness information, and adjusting an irradiation light source of the detection station according to the detection light source parameters; when the to-be-detected magnet arrives at the detection station, collecting a surface image of the to-be-detected magnet, and calculating a gray feature value of the surface image in real time; when the numerical deviation between the gray feature value and a preset gray target exceeds a preset deviation threshold value, finely adjusting the detection light source parameter according to the numerical deviation, and determining a target light source parameter; controlling the irradiation light source of the detection station to irradiate according to the target light source parameter, and obtaining the surface image of the to-be-detected magnet again as a target image; and inputting the target image into the AI detection model to obtain a defect detection result. By implementing the method, the detection effect on different magnets can be improved.
Owner:BEIJING HONGHAOXINDA TECH

Machine learning model training method, service data processing method, apparatus, and system

A machine learning model training method, a service data processing method, and an apparatus are provided, which are applied to the artificial intelligence field. In a training phase, a cloud server sends a machine learning submodel to an edge server. The edge server performs federated learning with client devices in a management domain of the edge server based on the obtained machine learning submodel, to obtain a trained machine learning submodel, and sends the trained machine learning submodel to the cloud server. The cloud server fuses obtained different trained machine learning submodels, to obtain a machine learning model. According to this application, training efficiency of the machine learning model can be improved. In an inference phase, the client device processes service data by using the trained machine learning submodel. According to this application, prediction efficiency of the machine learning model can be improved.
Owner:HUAWEI TECH CO LTD

Control of an industrial robot for a gripping task

In one aspect, the present invention relates to a distributed system for controlling at least one robot (R) in a gripping task for gripping objects (O) of different types which are arranged in a working area (CB, B, T) of the robot (R). The system comprises a central training computer (CTR), which is designed for pre- and post-training, and at least one local processing unit (LCU), on which real image data of the object (O) is recorded, which is used to generate post-training data.
Owner:VATHOS GMBH

Dynamic Latent Space Adaptation Based on Spatiotemporal Kernal Context for Multiscale Rendering

A system for dynamic latent space adaptation using spatiotemporal kernel context for multiscale rendering with hierarchical and Lorentzian autoencoders. The Spatiotemporal Kernel Estimator (SKE) analyzes media through motion field, temporal recurrence, frequency band, and scene semantics analyzers to generate adaptive kernel parameters encoding content-specific importance distributions. The system dynamically adapts latent manifold geometry by modifying metric tensor properties according to kernel context, enabling content-aware compression that allocates representational capacity based on visual significance. A multiscale cache implements kernel-adaptive retention policies prioritizing important regions. An adaptive renderer provides intelligent level-of-detail selection based on zoom level and kernel-estimated importance, optimizing processing allocation. The self-optimizing architecture continuously refines kernel context and geometric adaptation based on user interaction and performance feedback, achieving superior compression ratios and perceptual quality. Applications include bandwidth-efficient video streaming, virtual reality, scientific visualization, and cognitive video analytics requiring intelligent context-aware visual processing.
Owner:ATOMBEAM TECH INC

Gaze determination using one or more neural networks

Apparatuses, systems, and techniques are presented to predict gaze of an observer. In at least one embodiment, a network is trained to predict a gaze of one or more users based, at least in part, on one or more gazes corresponding to objects not always visible to the one or more users.
Owner:NVIDIA CORP

R-SpiNNaker System

An R-SpiNNaker system is provided. The R-SpiNNaker system provided includes a plurality of chips, one or more of which include a digital router, an analog signal router, and one or more network ports or point-to-point ports; the network port or point-to-point port of a first chip among the plurality of chips is coupled to the network port or point-to-point port of a second chip among the plurality of chips; the digital router of the first chip receives a data packet through the network port or point-to-point port of the chip where the digital router is located; and one or more switches of the analog signal router of the first chip are set according to data of a data packet carrying configuration information or programming information of the analog signal router.
Owner:北京他山科技股份有限公司

Training neural networks independently

Apparatuses, systems, and techniques to train neural networks. In at least one embodiment, a first neural network is trained to match accuracy of a second neural network independently of outputs of the second neural network based on, for example, a third neural network that generates weights for the first neural network.
Owner:NVIDIA CORP

Real-time data processing method and chip of novel computing low-power AI processor

The invention relates to the technical field of low-power-consumption AI processors, and discloses a novel real-time data processing method and chip for calculating a low-power-consumption AI processor, and the method comprises the steps: carrying out the neural mimicry classification processing of a calculation task inputted by the low-power-consumption AI processor, and obtaining a calculation task feature matrix; performing multi-dimensional mapping operation on the calculation task feature matrix to obtain a heterogeneous calculation space vector; performing core resource allocation based on the heterogeneous calculation space vector to obtain a dual-mode processing unit configuration matrix; constructing a storage and calculation integrated data path based on the dual-mode processing unit configuration matrix to obtain a parallel calculation data flow diagram; executing dynamic voltage frequency mapping on the parallel computing data flow diagram to obtain a power consumption optimization control vector; and inputting the power consumption optimization control vector into the trained pulse neural network model to carry out computing power load balancing calculation and generate computing power load balancing parameters, and the method improves the real-time data processing capability of the low-power AI processor.
Owner:SHENZHEN ITZR TECH

Inverted bottleneck architecture search and efficient attention mechanism for machine-learned models

Generally, the present disclosure is directed to efficient neural network architectures, techniques for constructing new efficient neural networks, and approaches to low-latency execution of neural networks. In an example aspect, the present disclosure provides a more powerful configuration of an inverted bottleneck block that maintains an efficient execution profile. An example search system can parameterize a universal inverted bottleneck block with a first parameter that categorically activates a spatial mixing operation in the unexpanded state. In this manner, for instance, a large variety of different network architectures can be explored using a relatively compact search space that admits high levels of parameter sharing. Further, the present disclosure introduces an efficiency-optimized multi-query attention block.
Owner:GOOGLE LLC

Efficiently serving machine-learned model computations with high throughput and low latency

An example method includes receiving input requests to process a plurality of input sequences using the machine-learned sequence processing model to generate a plurality of output sequences respectively corresponding to the plurality of input sequences; generating a plurality of initial attention tensors respectively for the plurality of input sequences, wherein: one or more respective initial attention tensors are generated for each respective input sequence in parallel over input elements of the respective input sequence; and the one or more respective initial attention tensors are generated in one or more batches having a first batch size using a prefill system that comprises one or more prefill computing devices and executes one or more layers of the machine-learned sequence processing model; and autoregressively generating, using the plurality of initial attention tensors, a plurality of output elements for each of the plurality of output sequences in one or more batches having a second batch size, wherein: the plurality of output elements are generated using a generation system that comprises one or more generation computing devices.
Owner:GOOGLE LLC

Clock signal distribution using photonic fabric

Various embodiments provide for clock signal distribution within a processor, such as a machine learning (ML) processor, using a photonic fabric.
Owner:SICILY MERGER SUB II INC

Urban road carbon emission prediction and optimization control method and system

The invention relates to an urban road carbon emission prediction and optimization control method and system, and belongs to the technical field of road traffic carbon emission monitoring, and the method comprises the steps: obtaining real-time monitoring data of a to-be-detected road, and carrying out the preprocessing; performing space-time alignment on the preprocessed real-time monitoring data, extracting dynamic traffic features and static road features, and generating a space-time feature matrix; inputting the spatial-temporal characteristic matrix into a pre-trained carbon emission prediction model to obtain a prediction result matrix of the to-be-detected road; on the basis of the prediction result matrix, identifying a road grid of which the carbon emission intensity prediction value exceeds a preset threshold value as a high emission area; acquiring real-time traffic state data of the high emission area; and based on the real-time traffic state data and the prediction result matrix of the high emission area, constructing a multi-objective optimization function and carrying out solving to obtain an optimization control strategy of the high emission area. According to the invention, traffic carbon emission optimization control and real-time traffic condition synchronization can be realized.
Owner:XIAN MUNICIPAL CONSTR GRP CO LTD

System and Method for Token-based Graphics Processing Unit (GPU) Utilization

A method, computer program product, and computing system for processing workload data associated with processing a plurality of requests for an artificial intelligence (AI) model on a processing unit. A maximum number of key-value (KV) cache blocks available for the workload data is determined by simulating the workload data using a simulation engine. A token utilization for the workload data is determined based upon, at least in part, the maximum number of KV cache blocks available for the workload data. Processing unit resources are allocated for the processing unit based upon, at least in part, the token utilization.
Owner:MICROSOFT TECHNOLOGY LICENSING LLC

Single-stage category-level object pose estimation

Apparatuses, systems, and techniques to determine a pose and relative dimensions of an object from an image. In at least one embodiment, a pose and relative dimensions of an object are determined from an image based at least in part on, for example, features of the image.
Owner:NVIDIA CORP

Fine adjustment and deployment method and system for domain-specific large model based on adaptive optimization

The invention discloses a field specialized large model fine tuning and deployment method and system based on adaptive optimization, belongs to the technical field of artificial intelligence and machine learning, and aims to solve the technical problem of how to improve the rapid adaptation capability of a large model on small-scale field data. In order to meet the requirements of special fields of law, medical treatment, finance and the like on professional terms and complex contexts, the technical scheme adopted by the invention comprises the following steps of: data processing and sample generation: performing cleaning, feature extraction and small sample expansion on field data, and generating a high-quality training sample through a field feature guide mechanism; small-sample fine tuning and migration: realizing rapid field adaptation of a large model through a small-sample learning technology and dynamic Prompt optimization, and reducing dependence on large-scale annotation data in combination with cross-field migration learning; compressing and optimizing the model; performing automatic deployment and collaborative reasoning; and monitoring and adaptive optimization are carried out.
Owner:INSPUR TIANYUAN COMM INFORMATION SYST CO LTD

Electro-photonic network for machine learning

Various embodiments provide for electro-photonic networks, including a plurality of processing elements connected by bidirectional photonic channels, suited for implementing neural-network models. Weights of the model may be preloaded into memory of the processing elements based on assignments of neural nodes to processing elements implementing them, and routers of the processing elements can be configured to stream activations between the processing elements based on a predetermined flow of activations in the model.
Owner:SICILY MERGER SUB II INC