3920 results about "Nerve network" patented technology

The invention discloses a vehicle type recognition method based on a rapid R-CNN deep neural network, which mainly comprises unsupervised deep learning, a multilayer CNN (Convolutional Neural Network), a regional advice network, network sharing and a softmax classifier. The vehicle type recognition method realizes a framework for implementing end-to-end vehicle detection and recognition by using one rapid R-CNN network in a real sense, and is capable of carrying out quick vehicle sub-category recognition with high accuracy and robustness under the environment of being applicable to the shape diversity, the illumination variation diversity, the background diversity and the like of vehicle targets.

The invention discloses a text named entity recognition method based on Bi-LSTM, CNN and CRF. The method includes the following steps: (1) using a convolutional nerve network to encode and convert information on text word character level to a character vector; (2) combining the character vector and word vector into a combination which, as an input, is transmitted to a bidirectional LSTM neural network to build a model for contextual information of every word; and (3) in the output end of the LSTM neural network, utilizing continuous conditional random fields to carry out label decoding to a whole sentence, and mark the entities in the sentence. The invention is an end-to-end model without the need of data pre-processing in the un-marked corpus with the exception of the pre-trained word vector, therefore the invention can be widely applied for statement marking of different languages and fields.

The present invention extends to methods, systems, and computer program products for recognizing input gestures. A neural network is trained using example inputs and backpropagation to recognize specified input patterns. Input gesture data is representative of movements in contact on a multi-touch input display surface relative to one or more axes over time. Example inputs used for training the neural network to recognize a specified input pattern can be created from sampling input gesture data for example input gestures known to represent the specified input pattern. Trained neural networks can subsequently be used to recognize input gestures that are similar to known input gestures as the specified input pattern corresponding to the known input gestures.

The present invention, in one embodiment, is directed to a method and system for accelerating and improving the history matching of a well bore and/or reservoir simulation model using a neural network. The neural network provides a correlation between the calculated history match error and a selected set of parameters that characterize the well bore and/or the reservoir. The neural network iteratively varies a selection and/or the value of the parameters to provide at least one set of history match parameters having a value that provides a minimum for the calculated history matching error.

An intrusion detection system (IDS) that uses application monitors for detecting application-based attacks against computer systems. The IDS implements application monitors in the form of a software program to learn and monitor the behavior of system programs in order to detect attacks against computer hosts. The application monitors implement machine learning algorithms to provide a mechanism for learning from previously observed behavior in order to recognize future attacks that it has not seen before. The application monitors include temporal locality algorithms to increased the accuracy of the IDS. The IDS of the present invention may comprise a string-matching program, a neural network, or a time series prediction algorithm for learning normal application behavior and for detecting anomalies.

A physical neural network is disclosed, which includes a connection network comprising a plurality of molecular conducting connections suspended within a connection gap formed between one or more input electrodes and one or more output electrodes. One or more molecular connections of the molecular conducting connections can be strengthened or weakened according to an application of an electric field across said connection gap. Thus, a plurality of physical neurons can be formed from said molecular conducting connections of said connection network. Additionally, a gate can be located adjacent said connection gap and which comes into contact with said connection network. The gate can be connected to logic circuitry which can activate or deactivate individual physical neurons among said plurality of physical neurons.

The invention provides a calculation apparatus and method for an accelerator chip accelerating a deep neural network algorithm. The apparatus comprises a vector addition processor module, a vector function value calculator module and a vector multiplier-adder module, wherein the vector addition processor module performs vector addition or subtraction and/or vectorized operation of a pooling layer algorithm in the deep neural network algorithm; the vector function value calculator module performs vectorized operation of a nonlinear value in the deep neural network algorithm; the vector multiplier-adder module performs vector multiplication and addition operations; the three modules execute programmable instructions and interact to calculate a neuron value and a network output result of a neural network and a synaptic weight variation representing the effect intensity of input layer neurons to output layer neurons; and an intermediate value storage region is arranged in each of the three modules and a main memory is subjected to reading and writing operations. Therefore, the intermediate value reading and writing frequencies of the main memory can be reduced, the energy consumption of the accelerator chip can be reduced, and the problems of data missing and replacement in a data processing process can be avoided.

Apparatus and methods for high-level neuromorphic network description (HLND) using tags. The framework may be used to define nodes types, define node-to-node connection types, instantiate node instances for different node types, and/or generate instances of connection types between these nodes. The HLND format may be used to define nodes types, define node-to-node connection types, instantiate node instances for different node types, dynamically identify and/or select network subsets using tags, and/or generate instances of one or more connections between these nodes using such subsets. To facilitate the HLND operation and disambiguation, individual elements of the network (e.g., nodes, extensions, connections, I/O ports) may be assigned at least one unique tag. The tags may be used to identify and/or refer to respective network elements. The HLND kernel may comprises an interface to Elementary Network Description.

The invention discloses a deep convolutional neural network-based lung cancer preventing self-service health cloud service system. The system comprises a convolutional neural network used for deep learning and training identification, a segmentation module which segments out a lung region from a CT image based on a full convolutional neural network, a deep convolutional neural network used for lung cancer diagnosis classification, and a self-service health cloud service platform used for performing early prevention and treatment according to an identified suspected lung cancer type. According to the system, the automation and intelligentization level of mobile internet-based lung cancer screening can be effectively improved, more citizens can know and participate in self-service health detection, assessment and guidance, the sensitivity, specificity and accuracy of early lung cancer screening and clinical diagnosis are improved, the lung cancer can be early discovered, early diagnosed and early treated, and the self-health management capability is enhanced.

Apparatus and methods for universal node design implementing a universal learning rule in a mixed signal spiking neural network. In one implementation, at one instance, the node apparatus, operable according to the parameterized universal learning model, receives a mixture of analog and spiking inputs, and generates a spiking output based on the model parameter for that node that is selected by the parameterized model for that specific mix of inputs. At another instance, the same node receives a different mix of inputs, that also may comprise only analog or only spiking inputs and generates an analog output based on a different value of the node parameter that is selected by the model for the second mix of inputs. In another implementation, the node apparatus may change its output from analog to spiking responsive to a training input for the same inputs.

The invention discloses a depth image human body joint positioning method based on a convolution nerve network. The method is characterized by comprising a training process and an identification process. The training process comprises the following steps: 1, inputting a training sample; 2, initializing a deep convolution nerve network and its parameters, wherein the parameters comprise a weight and a bias of each layer edge; and 3, by use of a forward algorithm and a backward algorithm, learning the parameters of the constructed convolution nerve network. The identification process comprises the following steps: 4, inputting a test sample; and 5, performing regression on the input test sample by use of the trained convolution nerve network to find positions of human body joints. According to the invention, by use of the deep convolution nerve network and large data, multiple challenges such as shielding, noise and the like can be resisted, and the accuracy is quite high; and at the same time, by means of parallel calculation, the effect of accurately positioning the human body joints in real time can be realized.

A method of building predictive statistical models provides a dedicated aggregation module for each transactional record source. Each aggregation module aggregates the transactional records using a neural network function to produce a scalar output which can then be input to a traditional modeling function, which may employ either logistic regression, neural network, or radial basis function techniques. The output of the aggregation modules can be saved, and updated aggregation values can be updated by processing new transaction records and combining the new transaction values with the previous output values using a blending function. Parameters of the neural network in the aggregation module may be calculated simultaneously with the parameters of the traditional modeling module.

Methods of detecting an object in an image using a convolutional neural-network-based architecture that processes multiple feature maps of differing scales from differing convolution layers within a convolutional network to create a regional-proposal bounding box. The bounding box is projected back to the feature maps of the individual convolution layers to obtain a set of regions of interest (ROIs) and a corresponding set of context regions that provide additional context for the ROIs. These ROIs and context regions are processed to create a confidence score representing a confidence that the object detected in the bounding box is the desired object. These processes allow the method to utilize deep features encoded in both the global and the local representation for object regions, allowing the method to robustly deal with challenges in the problem of object detection. Software for executing the disclosed methods within an object-detection system is also disclosed.

A mechanism for compiling a generative description of an inference task into a neural network. First, an arbitrary generative probabilistic model from the exponential family is specified (or received). The model characterizes a conditional probability distribution for measurement data given a set of latent variables. A factor graph is generated for the generative probabilistic model. Each factor node of the factor graph is expanded into a corresponding sequence of arithmetic operations, based on a specified inference task and a kind of message passing algorithm. The factor graph and the sequences of arithmetic operations specify the structure of a neural network for performance of the inference task. A learning algorithm is executed, to determine values of parameters of the neural network. The neural network is then ready for performing inference on operational measurements.

Efficient updates of connections in artificial neuron networks may be implemented. A framework may be used to describe the connections using a linear synaptic dynamic process, characterized by stable equilibrium. The state of neurons and synapses within the network may be updated, based on inputs and outputs to/from neurons. In some implementations, the updates may be implemented at regular time intervals. In one or more implementations, the updates may be implemented on-demand, based on the network activity (e.g., neuron output and/or input) so as to further reduce computational load associated with the synaptic updates. The connection updates may be decomposed into multiple event-dependent connection change components that may be used to describe connection plasticity change due to neuron input. Using event-dependent connection change components, connection updates may be executed on per neuron basis, as opposed to per-connection basis.

he invention discloses a medical information extraction system and method based on depth learning and distributed semantic features. The system is composed of a pretreatment module, a linguistic-model-based word vector training module, a massive medical knowledge base reinforced learning module, and a depth-artificial-neural-network-based medical term entity identification module. With a depth learning method, generation of the probability of a linguistic model is used as an optimization objective; and a primary word vector is trained by using medical text big data; on the basis of the massive medical knowledge base, a second depth artificial neural network is trained, and the massive knowledge base is combined to the feature leaning process of depth learning based on depth reinforced learning, so that distributed semantic features for the medical field are obtained; and then Chinese medical term entity identification is carried out by using the depth learning method based on the optimized statement-level maximum likelihood probability. Therefore, the word vector is generated by using lots of unmarked linguistic data, so that the tedious feature selection and optimization adjustment process during medical natural language process can be avoided.

A preemptive neural network database load balancer configured to observe, learn and predict the resource utilization that given incoming tasks utilize. Allows for efficient execution and use of system resources. Preemptively assigns incoming tasks to particular servers based on predicted CPU, memory, disk and network utilization for the incoming tasks. Direct write-based tasks to a master server and utilizes slave servers to handle read-based tasks. Read-base tasks are analyzed with a neural network to learn and predict the amount of resources that tasks will utilize. Tasks are assigned to a database server based on the predicted utilization of the incoming task and the predicted and observed resource utilization on each database server. The predicted resource utilization may be updated over time as the number of records, lookups, images, PDFs, fields, BLOBs and width of fields in the database change over time.

An apparatus and method for monitoring a process involve development and application of a statistically qualified neuro-analytic (SQNA) model to accurately and reliably identify process change. The development of the SQNA model is accomplished in two stages: deterministic model adaption and stochastic model modification of the deterministic model adaptation. Deterministic model adaption involves formulating an analytic model of the process representing known process characteristics, augmenting the analytic model with a neural network that captures unknown process characteristics, and training the resulting neuro-analytic model by adjusting the neural network weights according to a unique scaled equation error minimization technique. Stochastic model modification involves qualifying any remaining uncertainty in the trained neuro-analytic model by formulating a likelihood function, given an error propagation equation, for computing the probability that the neuro-analytic model generates measured process output. Preferably, the developed SQNA model is validated using known sequential probability ratio tests and applied to the process as an on-line monitoring system. Illustrative of the method and apparatus, the method is applied to a peristaltic pump system.

The invention discloses a Chinese question-answering system based on a neural network, which comprises a user interface module, a question word pre-segmentation module, a nerve cell pre-tagging module, a learning and training module, a nerve cell knowledge base module, a semantic block identification module, a question set index module and an answer reasoning module. The system comprises the steps of: firstly adopting an SIE encoding mode to encode the in-vocabulary words of the semantic block according to corresponding position, later converting an identification problem of the question semantic block into a tagging classification problem, and then adopting a classification model based on the neural network to determine the semantic structure of the question, and finally combing the semantic structure of the question to realize the question similarity computation based on the neural network and comparing the weight of various semantic features of the question by extracting the tagged semantic features of the question, thereby providing a basis for final answer reasoning. The Chinese question-answering system integrates the syntax, the semantics and the contextual knowledge of the question and can simulate the process that human beings process the sentence.

The invention provides a hardware accelerator and method for realizing a sparse GRU neural network based on FPGA. According to the invention, an apparatus for realizing sparse GRU neural network includes: an input receiving unit which is intended for receiving a plurality of input vectors and distributing the plurality of input vectors to a plurality of computing units; a plurality of computing units which acquire input vectors from the input receiving unit, read weight matrix data of a neural network, decoding the weight matrix data and conduct matrix calculation on the decoded weight matrix data and the input vectors, and output the result of the matrix calculation to a hidden layer state computing module; a hidden layer state computing module which acquires the result of matrix calculation from the computing units PE, and computing the state of the hidden layer; and a control unit which is intended for global control. In addition, the invention also provides a method for realizing sparse GRU neural network through iteration.

In Pavlovian and instrumental conditioning, rewards typically come seconds after reward-triggering actions, creating an explanatory conundrum known as the distal reward problem or the credit assignment problem. How does the brain know what firing patterns of what neurons are responsible for the reward if (1) the firing patterns are no longer there when the reward arrives and (2) most neurons and synapses are active during the waiting period to the reward? A model network and computer simulation of cortical spiking neurons with spike-timing-dependent plasticity (STDP) modulated by dopamine (DA) is disclosed to answer this question. STDP is triggered by nearly-coincident firing patterns of a presynaptic neuron and a postsynaptic neuron on a millisecond time scale, with slow kinetics of subsequent synaptic plasticity being sensitive to changes in the extracellular dopamine DA concentration during the critical period of a few seconds after the nearly-coincident firing patterns. Random neuronal firings during the waiting period leading to the reward do not affect STDP, and hence make the neural network insensitive to this ongoing random firing activity. The importance of precise firing patterns in brain dynamics and the use of a global diffusive reinforcement signal in the form of extracellular dopamine DA can selectively influence the right synapses at the right time.

A simple format is disclosed and referred to as Elementary Network Description (END). The format can fully describe a large-scale neuronal model and embodiments of software or hardware engines to simulate such a model efficiently. The architecture of such neuromorphic engines is optimal for high-performance parallel processing of spiking networks with spike-timing dependent plasticity. Neuronal network and methods for operating neuronal networks comprise a plurality of units, where each unit has a memory and a plurality of doublets, each doublet being connected to a pair of the plurality of units. Execution of unit update rules for the plurality of units is order-independent and execution of doublet event rules for the plurality of doublets is order-independent.

The invention discloses a human face detection method and a detection device based on multi-task cascade-connection convolution neural network, wherein the method comprises: establishing a cascade-connection multi-level convolution neural network; using the human face front samples, the human face back samples, some parts of the human face and the human face's key point samples as the training samples to train the multi-level convolution neural network to learn the tasks of human face categorizing, human face area position regression and human face's key point positioning; and utilizing the well trained multi-level convolution neural network to make human face detection from the to-be-detected image wherein in the training stage, both the online manner and the offline manner are combined to extract the human face back samples as the training samples. According to the invention, based on the cascade-connection multi-level convolution neural network, it is possible to learn the characteristics with stronger robustness, and at the same time, through the combination of the online manner and the offline manner to extract the back samples, the categorizing capability of the network is enhanced, so that the detection capability and the accuracy of the network are increased, and that the running speed of the method in an actual product is ensured.

We present a data-driven system that seamlessly integrates the compilation of a diverse set of patient electronic medical records, family history, and environmental risk factor databases, a program that determines correlations and assigns weights through a neural network between illness and the state of symptoms, a program to create disease-specific questionnaires and content, a program alert that informs users of trends and potential illness outbreaks due to anomalous changes in regional or national symptom trends, and a program that enables medical practitioners to view patient-generated records and manage an online referral system. The automated diagnoses and alerts allow users to make informed and rapid medical treatment decisions.

A method for creating an interpretable model for healthcare predictions includes training, by a deep learning processor, a neural network to predict health information by providing training data, including multiple combinations of measured or observed health metrics and corresponding medical results, to the neural network. The method also includes determining, by the deep learning processor and using the neural network, prediction data including predicted results for the measured or observed health metrics for each of the multiple combinations of the measured or observed health metrics based on the training data. The method also includes training, by the deep learning processor or a learning processor, an interpretable machine learning model to make similar predictions as the neural network by providing mimic data, including combinations of the measured or observed health metrics and corresponding predicted results of the prediction data, to the interpretable machine learning model.

The invention discloses a CT image pulmonary nodule detection system based on 3D full-connection convolution neural network. The detection system comprises the following five steps: constructing a training set data; performing 3D convolution neural network classification network training; performing 3D convolution neural network segmentation network training; carrying out false-positive suppression in which the trained segmentation network and the false-positive are utilized to inhibit the network; and detecting the pulmonary nodule. The technical schemes of the invention can realize the full automatic detection without any human intervention. At the same time, the recall rate of pulmonary nodule detection can be increased effectively; the false-positive focus of infection is reduced considerably; and a pixel level positioning quantitative and qualitative result for the focus-of-infection area of the pulmonary nodule can be obtained.

The invention discloses a multi-task layered image retrieval method based on a depth self-coding convolution neural network. The method is characterized by mainly comprising a multi-task end-to-end convolution neural network for deep learning and training recognition, a rapid visual segmentation detection and positioning method of a region-of-interest secondary screening module based on an RPN network, a coarse search of a full-graph sparse hash code, an area sensing semantic feature and matrix h accurate comparison and search based on the maximum response, and a region-of-interest selectivitycomparison algorithm. According to the method, the end-to-end training can be achieved, the interest region with higher quality can be automatically selected, the automation degree and the intelligent level of search by images can be effectively improved, and the image retrieval requirements of the big data age can be met by using little storage space at a high search speed.

Apparatus and methods for conditional plasticity in a neural network. In one approach, conditional plasticity mechanism is configured to select alternate plasticity rules when performing connection updates. The selection mechanism is adapted based on a comparison of actual connection efficiency and target efficiency. For instance, when actual efficiency is below the target value, the STDP rule may be modulated to increase long term potentiation. Similarly, when actual efficiency is above the target value, the STDP rule may be modulated to increase long term connection depression. The conditional plasticity mechanism dynamically adjusts connection efficacy, and prevents uncontrolled increase of connection weights, thereby improving network operation when processing information of a varying nature.

The invention discloses a video description method based on deep learning and text summarization. The video description method comprises the following steps: through a traditional image data set, training a convolutional neural network model according to an image classification task; extracting a video frame sequence of a video, utilizing the convolutional neural network model to extract convolutional neural network characteristics to form a <video frame sequence, text description sequence> pair which is used as the input of a recurrent neural network model, and training to obtain the recurrent neural network model; describing the video frame sequence of the video to be described through the recurrent neural network model obtained by training to obtain description sequences; and through a method that graph-based vocabulary centrality is used as the significance of the text summarization, sorting the description sequences, and outputting a final description result of the video. An event which happens in one video and object attributes associated with the event are described through a natural language so as to achieve a purpose that video contents are described and summarized.

An architecture for capture and generation, and a set of methods for characterization, prediction, and classification of traffic in packet networks are disclosed. The architecture consists of a device that stores packet timing information and processes the data so that characterization, prediction, and classification algorithms can perform operations in real-time. A methodology is disclosed for real-time traffic analysis, characterization, prediction, and classification in packet networks. The methodology is based on the simultaneous aggregation of packet arrival times at different times scales. The traffic is represented at the synchronous carrier level by the arrival or non-arrival of a packet. The invention does not require knowledge about the information source, nor needs to decode the information contents of the packets. Only the arrival timing information is required. The invention provides a characterization of the traffic on packet networks suitable for a real-time implementation. The methodology can be applied in real-time traffic classification by training a neural network from calculated second order statistics of the traffic of several known sources. Performance descriptors for the network can also be obtained by calculating the deviation of the traffic distribution from calculated models. Traffic prediction can also be done by training a neural network from a vector of the results of a given processing against a vector of results of the subsequent processing unit; noticing that the latter vector contains information at a larger time scale than the previous. The invention also provides a method of estimating an effective bandwidth measure in real time which can be used for connection admission control and dynamic routing in packet networks. The invention provides appropriate traffic descriptors that can be applied in more efficient traffic control on packet networks.