Abstract: Systems and methods for improving video understanding tasks based on higher-order object interactions (HOIs) between object features are provided. A plurality of frames of a video are obtained. A coarse-grained feature representation is generated by generating an image feature for each of for each of a plurality of timesteps respectively corresponding to each of the frames and performing attention based on the image features. A fine-grained feature representation is generated by generating an object feature for each of the plurality of timesteps and generating the HOIs between the object features. The coarse-grained and the fine-grained feature representations are concatenated to generate a concatenated feature representation.

Abstract: Methods and systems for training a neural network include sampling multiple local sub-networks from a global neural network. The local sub-networks include a subset of neurons from each layer of the global neural network. The plurality of local sub-networks are trained at respective local processing devices to produce trained local parameters. The trained local parameters from each local sub-network are averaged to produce trained global parameters.

Abstract: A computer-implemented method and computer processing system are provided. The method includes synchronizing, by a processor, respective ones of a plurality of data parallel workers with respect to an iterative distributed machine learning process. The synchronizing step includes individually continuing, by the respective ones of the plurality of data parallel workers, from a current iteration to a subsequent iteration of the iterative distributed machine learning process, responsive to a satisfaction of a predetermined condition thereby. The predetermined condition includes individually sending a per-receiver notification from each sending one of the plurality of data parallel workers to each receiving one of the plurality of data parallel workers, responsive to a sending of data there between. The predetermined condition further includes individually sending a per-receiver acknowledgement from the receiving one to the sending one, responsive to a consumption of the data thereby.

Abstract: A computer-implemented method and computer processing system are provided. The method includes synchronizing, by a processor, respective ones of a plurality of data parallel workers with respect to an iterative process. The synchronizing step includes individually continuing, by the respective ones of the plurality of data parallel workers, from a current iteration to a subsequent iteration of the iterative process, responsive to a satisfaction of a predetermined condition thereby. The predetermined condition includes individually sending a per-receiver notification from each sending one of the plurality of data parallel workers to each receiving one of the plurality of data parallel workers, responsive to a sending of data there between. The predetermined condition further includes individually sending a per-receiver acknowledgement from the receiving one to the sending one, responsive to a consumption of the data thereby.

Abstract: A network device, system, and method are provided. The network device includes a processor. The processor is configured to store a local estimate and a dual variable maintaining an accumulated subgradient for the network device. The processor is further configured to collect values of the dual variable of neighboring network devices. The processor is also configured to form a convex combination with equal weight from the collected dual variable of neighboring network devices. The processor is additionally configured to add a most recent local subgradient for the network device, scaled by a scaling factor, to the convex combination to obtain an updated dual variable. The processor is further configured to update the local estimate by projecting the updated dual variable to a primal space.

Abstract: A method for performing question answer (QA) tasks that includes entering an input into an encoder portion of an adaptive memory network, wherein the encoder portion parses the input into entities of text for arrangement of memory banks. A bank controller of the adaptive memory network organizes the entities into progressively weighted banks within the arrangement of memory banks. The arrangement of memory banks may be arranged to have an initial memory bank having lowest relevance for lowest relevance entities being closest to the encoder, and a final memory bank having a highest relevance for highest relevance entities being closes to a decoder. The method may continue with inferring an answer for the question answer (QA) task with the decoder analyzing only the highest relevance entities in the final memory bank.

Abstract: Systems and methods for improving video understanding tasks based on higher-order object interactions (HOIs) between object features are provided. A plurality of frames of a video are obtained. A coarse-grained feature representation is generated by generating an image feature for each of for each of a plurality of timesteps respectively corresponding to each of the frames and performing attention based on the image features. A fine-grained feature representation is generated by generating an object feature for each of the plurality of timesteps and generating the HOIs between the object features. The coarse-grained and the fine-grained feature representations are concatenated to generate a concatenated feature representation.

Abstract: Systems and methods for pruning a convolutional neural network (CNN) for surveillance with image recognition are described, including extracting convolutional layers from a trained CNN, each convolutional layer including a kernel matrix having at least one filter formed in a corresponding output channel of the kernel matrix, and a feature map set having a feature map corresponding to each filter. An absolute kernel weight is determined for each kernel and summed across each filter to determine a magnitude of each filter. The magnitude of each filter is compared with a threshold and removed if it is below the threshold. A feature map corresponding to each of the removed filters is removed to prune the CNN of filters. The CNN is retrained to generate a pruned CNN having fewer convolutional layers to efficiently recognize and predict conditions in an environment being surveilled.

Abstract: Systems and methods for predicting changes to an environment, including a plurality of remote sensors, each remote sensor being configured to capture images of an environment. A processing device is included on each remote sensor, the processing device configured to recognize and predict a change to the environment using a pruned convolutional neural network (CNN) stored on the processing device, the pruned CNN being trained to recognize features in the environment by training a CNN with a dataset and removing filters from layers of the CNN that are below a significance threshold for image recognition to produce the pruned CNN. A transmitter is configured to transmit the recognized and predicted change to a notification device such that an operator is alerted to the change.

Abstract: Systems and methods for surveillance are described, including an image capture device configured to mounted to an autonomous vehicle, the image capture device including an image sensor. A storage device is included in communication with the processing system, the storage device including a pruned convolutional neural network (CNN) being trained to recognize obstacles in a road according to images captured by the image sensor by training a CNN with a dataset and removing filters from layers of the CNN that are below a significance threshold for image recognition to produce the pruned CNN. A processing device is configured to recognize the obstacles by analyzing the images captured by the image sensor with the pruned CNN and to predict movement of the obstacles such that the autonomous vehicle automatically and proactively avoids the obstacle according to the recognized obstacle and predicted movement.

Abstract: A computer-implemented method and computer processing system are provided. The method includes synchronizing, by a processor, respective ones of a plurality of data parallel workers with respect to an iterative distributed machine learning process. The synchronizing step includes individually continuing, by the respective ones of the plurality of data parallel workers, from a current iteration to a subsequent iteration of the iterative distributed machine learning process, responsive to a satisfaction of a predetermined condition thereby. The predetermined condition includes individually sending a per-receiver notification from each sending one of the plurality of data parallel workers to each receiving one of the plurality of data parallel workers, responsive to a sending of data there between. The predetermined condition further includes individually sending a per-receiver acknowledgement from the receiving one to the sending one, responsive to a consumption of the data thereby.

Abstract: A method is provided for sparse communication in a parallel machine learning environment. The method includes determining a fixed communication cost for a sparse graph to be computed. The sparse graph is (i) determined from a communication graph that includes all the machines in a target cluster of the environment, and (ii) represents a communication network for the target cluster having (a) an overall spectral gap greater than or equal to a minimum threshold, and (b) certain information dispersal properties such that an intermediate output from a given node disperses to all other nodes of the sparse graph in lowest number of time steps given other possible node connections. The method further includes computing the sparse graph, based on the communication graph and the fixed communication cost. The method also includes initiating a propagation of the intermediate output in the parallel machine learning environment using a topology of the sparse graph.

Abstract: Systems and methods are disclosed for providing distributed learning over a plurality of parallel machine network nodes by allocating a per-sender receive queue at every machine network node and performing distributed in-memory training; and training each unit replica and maintaining multiple copies of the unit replica being trained, wherein all unit replicas train, receive unit updates and merge in parallel in a peer-to-peer fashion, wherein each receiving machine network node merges updates at later point in time without interruption and wherein the propagating and synchronizing unit replica updates are lockless and asynchronous.

Abstract: A context-aware attention-based neural network is provided for answering an input question given a set of purportedly supporting statements for the input question. The neural network includes a processing element. The processing element is configured to calculate a question representation for the input question, based on word annotations and word-level attentions calculated for the input question. The processing element is further configured to calculate a sentence representation for each of the purportedly supporting statements, based on word annotations and word-level attentions calculated for each of the purportedly supporting statements. The processing element is also configured to calculate a context representation for the set of purportedly supporting statements with respect to the sentence representation for each of the purportedly supporting statements.

Abstract: A computer-implemented method is provided for neural network training. The method includes improving a cache utilization by one or more processors during multiple training stages of a neural network, by performing a stage-wise mini-batch process on a set of samples used for the multiple training stages. The stage-wise mini-batch process waits for each of the multiple training stages to complete using a system wait primitive to improve the cache utilization.

Abstract: A face recognition system and method for face recognition are provided. The face recognition system includes a camera for capturing an input image of a face of a person to be recognized. The face recognition system further includes a cache. The face recognition system further includes a set of one or more processors configured to (i) improve a utilization of the cache by the one or more processors during multiple training stages of a neural network configured to perform face recognition, by performing a stage-wise mini-batch process on a set of samples used for the multiple training stages, and (ii) recognize the person by applying the neural network to the input image during a recognition stage. The stage-wise mini-batch process waits for each of the multiple training stages to complete using a system wait primitive to improve the utilization of the cache.

Abstract: Security systems and methods for detecting intrusion events include one or more sensors configured to monitor an environment. A pruned convolutional neural network (CNN) is configured process information from the one or more sensors to classify events in the monitored environment. CNN filters having the smallest summed weights have been pruned from the pruned CNN. An alert module is configured to detect an intrusion event in the monitored environment based on event classifications. A control module is configured to perform a security action based on the detection of an intrusion event.

Abstract: Methods and systems for pruning a convolutional neural network (CNN) include calculating a sum of weights for each filter in a layer of the CNN. The filters in the layer are sorted by respective sums of weights. A set of m filters with the smallest sums of weights is filtered to decrease a computational cost of operating the CNN. The pruned CNN is retrained to repair accuracy loss that results from pruning the filters.

Abstract: A computer-implemented method and computer processing system are provided. The method includes synchronizing, by a processor, respective ones of a plurality of data parallel workers with respect to an iterative process. The synchronizing step includes individually continuing, by the respective ones of the plurality of data parallel workers, from a current iteration to a subsequent iteration of the iterative process, responsive to a satisfaction of a predetermined condition thereby. The predetermined condition includes individually sending a per-receiver notification from each sending one of the plurality of data parallel workers to each receiving one of the plurality of data parallel workers, responsive to a sending of data there between. The predetermined condition further includes individually sending a per-receiver acknowledgement from the receiving one to the sending one, responsive to a consumption of the data thereby.

Abstract: Systems and methods for building a distributed learning framework, including generating a sparse communication network graph with a high overall spectral gap. The generating includes computing model parameters in distributed shared memory of a cluster of a plurality of worker nodes; determining a spectral gap of an adjacency matrix for the cluster using a stochastic reduce convergence analysis, wherein a spectral reduce is performed using a sparse reduce graph with a highest possible spectral gap value for a given network bandwidth capability; and optimizing the communication graph by iteratively performing the computing and determining until a threshold condition is reached. Each of the plurality of worker nodes is controlled using tunable approximation based on available bandwidth in a network in accordance with the generated sparse communication network graph.