Abstract: A hardware implementation of a neural network and a method of processing data in such a hardware implementation are disclosed. Input data for a plurality of layers of the network is processed in blocks, to generate respective blocks of output data. The processing proceeds depth-wise through the plurality of layers, evaluating all layers of the plurality of layers for a given block, before proceeding to the next block.

Abstract: Various embodiments of the present invention provide methods, apparatus, systems, computing devices, computing entities, and/or the like for performing predictive structural analysis. Certain embodiments of the present invention utilize systems, methods, and computer program products that perform predictive structural analysis using at least one of table column classification machine learning models, table column clustering machine learning models, structural variance generation machine learning models, and emergence report generation machine learning models.

Abstract: A device and a computer-implemented method for machine learning. First input data are provided which encompass information concerning dimensions and options for the machine learning. At least one of the options is associated with at least one of the dimensions as a function of information concerning the dimensions and options for at least one test case for the machine learning. A combination of options for a subset of the dimensions that is lacking in the set of test cases is determined, and a test case is determined for this combination.

Abstract: A computer-implemented method for training a neural network, which, in particular, is configured to classify physical measuring variables. The neural network is trained with the aid of a training data set. Pairs including an input signal and an associated desired output signal are drawn from the training data set for training. An adaptation of parameters of the neural network occurs as a function of an output signal of the neural network, when the input signal is supplied, and as a function of the desired output signal. The drawing of pairs always takes place from the entire training data set.

Abstract: A plurality of tensors is obtained, and the plurality of tensors is reformatted to provide a plurality of reformatted tensors of a select dimension. The reformatting includes adding padding to at least one reformatted tensor of the plurality of reformatted tensors. The plurality of reformatted tensors is concatenated to provide a concatenated tensor. The concatenated tensor is to be used in recurrent neural network processing.

Abstract: Embodiments for learning semantic description of data based on physical knowledge in a computing environment by a processor. Physical knowledge data and semantic labels associated with data from one or more data sources may be learned. Source attributes of the one or more data sources may be associated with one or more classes and concepts of a plurality of ontologies based on the physical knowledge data and the semantic labels to generate textual descriptors of the data.

Abstract: A building system of a building including one or more memory devices configured to store one or more instructions that, when executed on one or more processors, cause the one or more processors to exercise a building entity causing building entity data to be generated associated with the building entity, the building entity data indicating a result of exercising the building entity and collect the building entity data. The instructions cause the one or more processors to identify, based on a relational model, one or more relationships between one or more building entities and the building entity, wherein the one or more relationships indicate that exercising the building entity affects operation of the one or more building entities and identify that the building is experiencing a performance issue by analyzing the building entity data and the one or more relationships.

Abstract: In an approach for policy security shifting left of infrastructure as code compliance, a processor trains a neural network model to classify a code per policy and provide a policy vector score for the code associated with one or more policies. A processor enables the neural network model to scan and score a new code during a continuous integration and continuous deployment pipeline. A processor outputs a scanned score of the new code to a user. A processor retrains the neural network model by capturing a continuous integration and continuous deployment change and run-time compliance posture that occurs as a response by the user.

Abstract: An apparatus and method for generating user-specific self-executing data structures are described. The apparatus includes at least a processor and a memory communicatively coupled to the at least a processor. The memory includes instructions configuring the at least a processor to receive a user profile comprising a plurality of user related data associated with a user, analyze the plurality of user related data, determine at least one user designation associated with the user as a function of the analyzing the plurality of user related data, and generate a self-executing record as a function of the user designation for the user.

Abstract: An example system includes a processor to learn a shared embedding space on unlabeled videos using speech visual correspondence. The processor can learn a number of additional embeddings including a question plus video embedding and an answer embedding using the shared embedding space to generate a trained visual question answering model. The processor can execute a visual question answering based on the trained visual question answering model.

Abstract: Methods and systems for training a neural network include training language-specific teacher models using different respective source language datasets. A student model is trained, using the different respective source language datasets and soft labels generated by the language-specific teacher models, including shuffling the source language datasets and shuffling weights of language-dependent layers in language-specific parts of the student model. Weights of language-independent layers of the student model are copied to a language-independent layers of a target model to initialize language-independent layers of the target model. The target model is trained with a target language dataset.

Abstract: A machine learning system for evaluating at least one characteristic of a heart valve, an inflow tract, an outflow tract or a combination thereof may include a training mode and a production mode. The training mode may be configured to train a computer and construct a transformation function to predict an unknown anatomical characteristic and/or an unknown physiological characteristic of a heart valve, inflow tract and/or outflow tract, using a known anatomical characteristic and/or a known physiological characteristic the heart valve, inflow tract and/or outflow tract. The production mode may be configured to use the transformation function to predict the unknown anatomical characteristic and/or the unknown physiological characteristic of the heart valve, inflow tract and/or outflow tract, based on the known anatomical characteristic and/or the known physiological characteristic of the heart valve, inflow tract and/or outflow tract.

Abstract: A neural network computation apparatus includes a first processing block including a plurality of processing units that each perform a matrix multiplication operation on input data and weights, and a second processing block including a plurality of element-wise operation processing groups. The element-wise operation processing group selectively perform a first neural network computation operation and a second neural network computation operation. The first neural network computation operation comprises the matrix multiplication operation on the input data and the weights and an activation operation on a result value of the matrix multiplication operation, and the second neural network computation operation comprises an activation operation on the result value of the matrix multiplication operation, which is transferred from the first processing block, and an element-wise operation.

Abstract: Systems and methods are provided for analog hardware realization of neural networks. The method incudes obtaining a neural network topology and weights of a trained neural network. The method also includes transforming the neural network topology into an equivalent analog network of analog components. The method also includes computing a weight matrix for the equivalent analog network based on the weights of the trained neural network. Each element of the weight matrix represents a respective connection between analog components of the equivalent analog network. The method also includes generating a schematic model for implementing the equivalent analog network based on the weight matrix, including selecting component parameter values for the analog components.

Abstract: An example system includes a processor to receive a data set. The processor can generate a data slice rule based on a data observation for a data point in the data set. The processor can generate an instance of data based on the generated data slice rule.

Abstract: A general matrix-matrix (GEMM) accelerator core includes first and second buffers, a control logic circuit, and a first processing element (PE). The first buffer receives a elements of a first matrix A of activation values. The second buffer receives b elements of a second matrix B of weight values. The control logic circuit replaces a zero-valued a element in a first column of the first buffer with a nonzero-valued a element that is within a maximum borrowing distance of a location of the zero-valued a element in the first column of the first buffer. The PE receives a elements from the first column of the first buffer including the nonzero-valued element a selected to replace the zero-valued a element and receives b elements from locations in the second buffer that correspond to locations in the first buffer from where the a elements have been received by the PE.

Abstract: A feature engineering application receives a plurality of data sets from different data sources for training a model for making a prediction based on new data. The feature engineering application generates primitives based on the data sets. A primitive is to be applied to a variable in the data sets to synthesize a feature. The feature engineering application also receives a temporal parameter that specifies a temporal value for generating time-based features. After the primitives are generated and the temporal parameter is received, the feature engineering application aggregates the plurality of data entities based on primary variables in the plurality of data entities and generate an entity set based on the aggregation. The feature engineering application then synthesize features, including the time-based features, based on the entity set, at least some of the primitives, and the temporal parameter.

Abstract: A method includes obtaining, using at least one processor of an electronic device, one or more instance level supervised artificial intelligence (AI) models. The method also includes obtaining, using the at least one processor, aggregated level label information related to the one or more instance level supervised AI models. The method further includes obtaining, using the at least one processor, instance level feature information related to the one or more instance level supervised AI models. In addition, the method includes training, using the at least one processor, the one or more instance level supervised AI models using the instance level feature information and the aggregated level label information to obtain one or more trained instance level supervised AI models.

Abstract: The instant system and methods solves the cold start problem through various systems and methods directed to aggregating user interaction data associated with a user over a period of time, generating an embedding model based on the aggregated user interaction data, generating a content embedding vector based on the embedding model, generating an embedding profile vector based on the embedding model, storing the embedding profile vector in a storage device, receiving each of the content embedding vector and embedding vector profile for training a ranking model, and generating a predicted list of one or more content items of interest for recommending to the user.

Abstract: A framework is presented that provides a shift in the conceptual and practical realization of privacy-preserving interference on deep neural networks. The framework leverages the concept of the binary neural networks (BNNs) in conjunction with the garbled circuits protocol. In BNNs, the weights and activations are restricted to binary (e.g., ±1) values, substituting the costly multiplications with simple XNOR operations during the inference phase. The XNOR operation is known to be free in the GC protocol; therefore, performing oblivious inference on BNNs using GC results in the removal of costly multiplications. The approach consistent with implementations of the current subject matter provides for oblivious inference on the standard DL benchmarks being performed with minimal, if any, decrease in the prediction accuracy.