Abstract: This disclosure relates generally relates to method and system to process asynchronous and distributed training tasks. Training a large-scale deep neural network (DNN) model with large-scale training data is time-consuming. The method creates a work queue (Q) with a set of predefined number of tasks comprising a training data. Here, set of central processing units (CPUs) information and a set of graphics processing units (GPUs) information are fetched from the current environment to initiate a parallel process asynchronously on the work queue (Q) to train a set of deep learning models with optimized resources using a data pre-processing technique, to compute a transformed training data and training by using an asynchronous model training technique, the set of deep learning models on each GPU asynchronously with the transformed training data based on a set of asynchronous model parameters.

Abstract: State of art approaches independently use a Pruning-weight Clustering-Quantization (PCQ) or Knowledge Distillation (KD) for model optimization and require critical manual intervention. Embodiments of the present disclosure provide a method and system for the two-step hierarchical model optimization approach for generating optimized model DL model. The method comprises a AutoPCQ technique followed by conditional application of an automated KD (AKD) technique. The AutoPCQ technique formulates a problem of configuration selection of the DL model as an optimization problem by iteratively applying Bayesian optimization and Reinforcement Learning. Further, the AKD technique formulates automated search of a student model as the optimization problem with the DL model representing a teacher model. A search space for the student model is defined by a restricted Neural Network Architecture Search that restricts the search space.

Abstract: Machine vision devices may be configured to automatically connect to a remote management server (e.g., a “cloud”-based management server), and may offload and/or communicate images and analyses to the remote management server via wired or wireless communications. The machine vision devices may further communicate with the management server, user computing devices, and/or human machine interface devices, e.g., to provide remote access to the machine vision device, provide real-time information from the machine vision device, receive configurations/updates, provide interactive graphical user interfaces, and/or the like.

Abstract: The present disclosure describes methods, systems, and non-transitory computer-readable media for implementing a machine learning framework to generate a recommend digital assets from a digital image. For example, in one or more embodiments, the disclosed systems utilize a machine learning model to detect a shape, color, pattern, or other digital asset type from a digital image and then extract (and further modify) the detected asset type to create various different digital assets as recommendations. In some cases, the disclosed system utilizes the machine learning model to determine one or more digital asset classes associated with the digital image, generate preprocessed digital assets from the digital image for those digital asset classes, and generate production-ready digital assets from the preprocessed digital assets. Further, in some instances, the disclosed systems provide one or more of the digital assets via recommendations based on asset scores determined via the generation process.

Abstract: State of the art predictive maintenance systems that generate predictions with respect to maintenance of High Performance Computing (HPC) systems have the disadvantage that they either are reactive, or the predictions are affected due to quality issues associated with the data being collected from the HPC systems. The disclosure herein generally relates to predictive maintenance, and, more particularly, to a method and system for predictive maintenance of High Performance Computing (HPC) systems. The system performs abstraction and cleansing on performance data collected from the HPC systems, and generates a cleansed performance data, on which a Machine Leaning (ML) prediction is applied to generate predictions with respect to maintenance of the HPC systems.

Abstract: Controller devices may be configured to automatically connect to a remote management server (e.g., a “cloud”-based management server), and may offload received data and analyses to the remote management server via wired or wireless communications. The controller devices may further communicate with the management server, user computing devices, and/or human machine interface devices, e.g., to provide remote access to the controller device, provide real-time information from the controller device, receive configurations/updates, provide interactive graphical user interfaces, and/or the like.

Abstract: Controller devices may be configured to automatically connect to a remote management server (e.g., a “cloud”-based management server), and may offload received data and analyses to the remote management server via wired or wireless communications. The controller devices may further communicate with the management server, user computing devices, and/or human machine interface devices, e.g., to provide remote access to the controller device, provide real-time information from the controller device, receive configurations/updates, provide interactive graphical user interfaces, and/or the like.

Abstract: Machine vision devices may be configured to automatically connect to a remote management server (e.g., a “cloud”-based management server), and may offload and/or communicate images and analyses to the remote management server via wired or wireless communications. The machine vision devices may further communicate with the management server, user computing devices, and/or human machine interface devices, e.g., to provide remote access to the machine vision device, provide real-time information from the machine vision device, receive configurations/updates, provide interactive graphical user interfaces, and/or the like.

Abstract: Techniques are disclosed for assigning a geographical location to a photograph of a landscape. A geographic location tag is calculated from raw sensor data and a three-dimensional terrain map that corresponds to the landscape in the photograph. One technique includes calculating a two-dimensional ideal view projection of a camera used to take the landscape photograph based at least in part on a geographic location of the camera when the landscape photograph was taken and a field of view of the camera when the landscape photograph was taken. At least one of the topographical features visible from the camera based on the two-dimensional ideal view projection is selected from a three-dimensional terrain map representing a plurality of topographical features within a geographic region. At least one geographic location tag representing at least one of the selected topographical features is assigned to the landscape photograph.

Abstract: Controller devices may be configured to automatically connect to a remote management server (e.g., a “cloud”-based management server), and may offload received data and analyses to the remote management server via wired or wireless communications. The controller devices may further communicate with the management server, user computing devices, and/or human machine interface devices, e.g., to provide remote access to the controller device, provide real-time information from the controller device, receive configurations/updates, provide interactive graphical user interfaces, and/or the like.

Abstract: The disclosure generally relates to principal component analysis (PCA) computation and, more particularly, to concurrent PCA computation. In one embodiment, a plurality of concurrent PCA requests are received by a server. An input matrix for each of the concurrent PCA requests is computed using a general purpose-graphical processing unit (GP-GPU) by the server. Further, tridiagnolization on the input matrix is performed on each of the concurrent PCA requests by a general purpose-graphical processing unit (GP-GPU) in the server to generate a tridiagonal matrix for each of the concurrent PCA requests. Furthermore, a plurality of eigen values and corresponding eigen vectors are computed for the tridiagonal matrix of each of the concurrent PCA requests by the server and subsequently back transformation of the eigen values and the eigen vectors is performed by the server for each of the concurrent PCA requests to obtain associated principal components.

Abstract: This disclosure relates generally to correlation filters, and more particularly to designing of correlation filter. In one embodiment, a system for designing a correlation filter in a multi-processor system includes a multi-core processor coupled to a first memory and one or more co-processors coupled to one or more respective second memories. The multi-core processor partitions each of a plurality of frames associated with media content into a plurality of pixel-columns, and systematically stores said pixel-columns width-wise in a plurality of temporary matrices by a plurality of threads of the multi-core processor. The plurality of temporary matrices are transferred by the multi-core processor to one or more respective second memories in a plurality of streams simultaneously in an asynchronous mode. A plurality of filter harmonics of the correlation filter are computed by performing compute operations involving at least the plurality of temporary matrices, to obtain the correlation filter.

Abstract: Techniques are disclosed for assigning a geographical location to a photograph of a landscape. A geographic location tag is calculated from raw sensor data and a three-dimensional terrain map that corresponds to the landscape in the photograph. One technique includes calculating a two-dimensional ideal view projection of a camera used to take the landscape photograph based at least in part on a geographic location of the camera when the landscape photograph was taken and a field of view of the camera when the landscape photograph was taken. At least one of the topographical features visible from the camera based on the two-dimensional ideal view projection is selected from a three-dimensional terrain map representing a plurality of topographical features within a geographic region. At least one geographic location tag representing at least one of the selected topographical features is assigned to the landscape photograph.

Abstract: A computer implemented method and apparatus for extracting localizable content from an article. The method comprises accessing a page content tree comprising a plurality of nodes; generating a set of all properties for each node in the plurality of nodes of the page content tree; accessing a localization tree comprising a global set of localizable properties of nodes; generating a set of localizable properties from the localization tree for nodes that match nodes of the page content tree; performing an intersection of the set of all properties for each node of the page content tree and the set of localizable properties; and extracting a value for each property in the intersection.

Abstract: The disclosure generally relates to principal component analysis (PCA) computation and, more particularly, to concurrent PCA computation. In one embodiment, a plurality of concurrent PCA requests are received by a server. An input matrix for each of the concurrent PCA requests is computed using a general purpose-graphical processing unit (GP-GPU) by the server. Further, tridiagnolization on the input matrix is performed on each of the concurrent PCA requests by a general purpose-graphical processing unit (GP-GPU) in the server to generate a tridiagonal matrix for each of the concurrent PCA requests. Furthermore, a plurality of eigen values and corresponding eigen vectors are computed for the tridiagonal matrix of each of the concurrent PCA requests by the server and subsequently back transformation of the eigen values and the eigen vectors is performed by the server for each of the concurrent PCA requests to obtain associated principal components.

Abstract: This disclosure relates generally to correlation filters, and more particularly to designing of correlation filter. In one embodiment, a system for designing a correlation filter in a multi-processor system includes a multi-core processor coupled to a first memory and one or more co-processors coupled to one or more respective second memories. The multi-core processor partitions each of a plurality of frames associated with media content into a plurality of pixel-columns, and systematically stores said pixel-columns width-wise in a plurality of temporary matrices by a plurality of threads of the multi-core processor. The plurality of temporary matrices are transferred by the multi-core processor to one or more respective second memories in a plurality of streams simultaneously in an asynchronous mode. A plurality of filter harmonics of the correlation filter are computed by performing compute operations involving at least the plurality of temporary matrices, to obtain the correlation filter.

Abstract: This invention relates generally to electrode materials, electrochemical cells employing such materials, and methods of synthesizing such materials. The electrode materials have a crystal structure with a high ratio of Li to metal M, which is found to improve capacity by enabling the transfer of a greater amount of lithium per metal, and which is also found to improve stability by retaining a sufficient amount of lithium after charging. Furthermore, synthesis techniques are presented which result in improved charge and discharge capacities and reduced particle sizes of the electrode materials.

Abstract: The present invention generally relates to carbophosphates and other compounds. Such compounds may be used in batteries and other electrochemical devices, or in other applications such as those described herein. One aspect of the invention is generally directed to carbophosphate compounds, i.e., compounds containing carbonate and phosphate ions. For example, according to one set of embodiments, the compound has a formula Ax(M)(PO4)a(CO3)b, where M comprises one or more cations. A may include one or more alkali metals, for example, lithium and/or sodium. In some cases, x is greater than about 0.1, a is between about 0.1 and about 5.1, and b is between about 0.1 and about 5.1. In certain embodiments, the compound may have a unit cell atomic arrangement that is isostructural to unit cells of the minerals sidorenkite, bonshtedtite, bradleyite, crawfordite, or ferrotychite.

Abstract: A computer implemented method and apparatus for extracting localizable content from an article. The method comprises accessing a page content tree comprising a plurality of nodes; generating a set of all properties for each node in the plurality of nodes of the page content tree; accessing a localization tree comprising a global set of localizable properties of nodes; generating a set of localizable properties from the localization tree for nodes that match nodes of the page content tree; performing an intersection of the set of all properties for each node of the page content tree and the set of localizable properties; and extracting a value for each property in the intersection.

Abstract: In general, the invention relates to electrode materials, e.g., novel cathode materials with high density, low cost, and high safety. A voltage design strategy based on the mixing of different transition metals in crystal structures known to be able to accommodate lithium in insertion and delithiation is presented herein. By mixing a metal active on the +2/+3 couple (e.g., Fe) with an element active on the +3/+5 or +3/+6 couples (e.g., V or Mo), high capacity multi-electron cathodes are designed in an adequate voltage window.