Abstract: An enhanced dynamic random access memory (eDRAM)-based computing-in-memory (CIM) convolutional neural network (CNN) accelerator comprises four P2ARAM blocks, where each of the P2ARAM blocks includes a 5T1C ping-pong eDRAM bit cell array composed of 64×16 5T1C ping-pong eDRAM bit cells. In each of the P2ARAM blocks, 64×2 digital time converters convert a 4-bit activation value into different pulse widths from a row direction and input the pulse widths into the 5T1C ping-pong eDRAM bit cell array for calculation. A total of 16×2 convolution results are output in a column direction of the 5T1C ping-pong eDRAM bit cell array. The CNN accelerator uses the 5T1C ping-pong eDRAM bit cells to perform multi-bit storage and convolution in parallel. An S2M-ADC scheme is proposed to allot an area of an input sampling capacitor of an ABL to sign-numerical SAR ADC units of a C-DAC array without adding area overhead.

Abstract: Disclosed herein are systems and methods for determining data structures. In some embodiments, a classifier may be used to determine one or more attributes of an entity. In some embodiments, a clustering algorithm may be used to determine an attribute cluster. In some embodiments, an impact metric machine learning model may be used to determine an outlier cluster. In some embodiments, an outlier process may be determined as a function of the outlier cluster. In some embodiments, a visual element may be determined as a function of an outlier process and may be displayed to a user.

Abstract: A system, method and computer program product for interfacing a decision engine and a marketing engine in order to provide vendor-related data in response to decision-related data is disclosed. In at least one embodiment, the system and method may include providing a decision engine on a user-accessible network, interfacing a marketing engine with the decision engine on the network; receiving a plurality of user inputs with the decision engine; processing decision-related data with the decision engine in accordance with the plurality of user inputs; sharing the decision-related data with the marketing engine; processing the decision-related data with the marketing engine; and transmitting vendor-related data via the network.

Abstract: Techniques for deriving additional features from input data are described herein. Input data from a plurality of source files are received. One or more features corresponding to the input data, which includes information about semantic types, is identified. The input data is then processed to generate additional features for the input data. New data corresponding to the additional features are then generated and access to the new data is subsequently provided.

Abstract: An agricultural monitoring system, apparatus and method(s) for providing crop-related forecasts by performing the steps of receiving seasonal image data from at least one source, where the seasonal image data is associated with at least one agricultural field, processing the seasonal image data using a Bayesian framework, where the Bayesian framework comprises one or more crop models configured to predict, based on the seasonal image data, one or more probabilities indicative of at least one crop state, updating at least one crop model of the Bayesian framework based on the one or more probabilities, and outputting a forecast of the at least one crop state based on the one or more probabilities.

Abstract: A layered machine learning system for processing data. The machine learning system comprises decision trees with different depths. An iterative training process is performed on the layered machine learning system to determine the structures of the decision trees based on prior predictions. The fitted decision trees are further configured to update leaf values with a gradient boosting method. By cumulating the predictions of decisions trees in prior iterations, interaction effects are modeled among different depths within the layered machine learning system.

Abstract: A dynamic reasoning system may include a symbolic reasoning engine that iteratively calls a dynamic rule generator to answer an input query. The symbolic reasoning engine may determine a primary goal and/or secondary goals to generate proofs for the answer. The symbolic reasoning engine may call a rules component to provide rules to prove a current input goal. The rules component may use a static rule knowledge base and/or the dynamic rule generator to retrieve and rank rules relevant to the current input goal. The dynamic rule generator may generate new rules that lead to the current input goal. The dynamic rule generator may include a statistical model that generates unstructured or structured probabilistic rules based on context related to the input query. The symbolic reasoning engine may return a list of rules with confidence for explaining the answer to the input goal.

Abstract: A method, a device, and a non-transitory storage medium each provide an object recognition service that identifies an object within an image or a video. An object is identified according to a general classification based on a domain-based inference model, and is subsequently further identified according to a sub-classification of the general classification based on another domain-based inference model. The domain-based inference models are hierarchical. The object recognition of the object may be used in support of end user services.

Abstract: Original idea extraction from written data is provided by capturing expression as written text data, obtaining a knowledge graph representing concepts and relationships between the concepts automatically topic modeling the written text data to ascertain thought units and identify respective concepts of the thought unit, mapping a thought unit to the knowledge graph, determining that the thought unit is an original idea, and based on determining that the thought unit is an original idea, storing a representation of the original idea to an idea repository and invoking processing of at least one computer.

Abstract: Certain aspects of the present disclosure provide a method for performing quantum convolution, including: receiving input data at a neural network model, wherein the neural network model comprises at least one quantum convolutional layer; performing quantum convolution on the input data using the at least one quantum convolutional layer; generating an output wave function based on the quantum convolution using the at least one quantum convolution layer; generating a marginal probability distribution based on the output wave function; and generating an inference based on the marginal probability distribution.

Abstract: In variants, a method for analog product determination can include: determining functional property feature values for a target and determining variable values for a prototype based on the functional property feature values for the target.

Abstract: The disclosed embodiments relate to a system that optimizes execution of a DNN based on operational performance parameters. During operation, the system collects the operational performance parameters from the DNN during operation of the DNN, wherein the operational performance parameters include parameters associated with operating conditions for the DNN, parameters associated with resource utilization during operation of the DNN, and parameters associated with accuracy of results produced by the DNN. Next, the system uses the operational performance parameters to update the DNN model to improve performance and efficiency during execution of the DNN.

Abstract: An information processing device, includes a memory; and a processor coupled to the memory and configured to: calculate a quantization error when a variable to be used in a neural network is quantized, generate a threshold value based on reference information related to a first recognition rate obtained by past learning of the neural network and a second recognition rate that is obtained by calculation of the neural network, determine a variable of data type to be quantized among variables to be used for calculation of the neural network based on the calculated quantization error and the generated threshold value, and execute the calculation of the neural network by using the variable of data type.

Abstract: An agent selection component that selects and activates artificial agents for interacting with a user on behalf of various entity based on user interaction with a real environment. The agent selection component detects user interaction with the real environment, and evaluates the detected user interaction. Based on that evaluation, the agent selection component selects an artificial agent that acts for an entity from amongst multiple artificial agents that act for different entities (e.g., to answer questions, to place orders, to schedule, or the like). The agent selection component then causes the selected artificial agent to activate to interact with the entity. Thus, different interactions with a real environment may result in the agent selection component permitting the user to interface with artificial agents for different entities.

Abstract: In a system for enabling configuration of an ensemble of several solvers, such that the ensemble can efficiently solve a constraint problem, for each one of several candidate configurations, an array of scores is computed. The array corresponds to a statistical parameter related to a problem solution, and the computation is based on, at least in part, a set of features associated with the problem. One candidate configuration is assigned to a solver, and based on the array of scores associated with that candidate configuration the same or a different candidate configuration is assigned to a another solver. A system for dynamically reconfiguring an ensemble of solvers obtains runtime data from several solvers, and a new configuration is determined by applying a machine learning and/or heuristic analysis procedure to the runtime data. The configuration of a solver may be updated according to the new configuration while that solver is running.

Abstract: An online system performs predictions for real-time tasks and near real-time tasks that need to be performed by a deadline. A client device receives a real-time machine learning based model associated with a measure of accuracy. If the client device determines that a task can be performed using predictions having less than the specified measure of accuracy, the client device uses the real-time machine learning based model. If the client device determines that a higher level of accuracy of results is required, the client device sends a request to an online system. The online system provides a prediction along with a string representing a rationale for the prediction.

Abstract: A method for providing real-time recommendations to user in a shopping environment involves sampling a shopping environment using video cameras to generate video features related to a shopper in connection to an item, the sampling input to a machine learning model to create labels related to a state of a scenario, the scenario including the shopper handling the item. Supplemental information is provided to the shopper in connection with the item. The makeup of the supplemental information may be sourced from online service or device associated with the shopper. The supplemental information may be delivered to the shopper, or to a shopper-aware display in the store, or elsewhere. A processing entity associated with the store detects a scenario to identify the shopper as having finished shopping, and causing a charge of the item to a cashierless shopping cart associated with the shopper.

Abstract: In example embodiments, an enhanced deep belief learning model with an extended Kalman filter (EKF) is used for training and updating a deep belief network (DBN) with new data to produce a DBN model useful in making predictions on a variety of types of datasets, including data captured from infrastructure-attached sensors describing the condition of the infrastructure. The EKF is employed to estimate operation parameters of the DBN and generate the model's output covariance. Further, in example embodiments, the configuration of the DBN model may be optimized by a competent genetic algorithm.

Abstract: Methods, devices, and computer-readable media for multi-task based lifelong learning. A method for lifelong learning includes identifying a new task for a machine learning model to perform. The machine learning model trained to perform an existing task. The method includes adaptively training a network architecture of the machine learning model to generate an adapted machine learning model based on incorporating inherent correlations between the new task and the existing task. The method further includes using the adapted machine learning model to perform both the existing task and the new task.

Abstract: Technologies for a quantum/classical hybrid approach to solving optimization problems is disclosed. In the illustrative embodiment, an optimization problem is decomposed into two sub-problems. The first sub-problem is solved on a classical computer, and a result from the first sub-problem is provided to a quantum computer. The quantum computer then solves the second sub-problem based on the result of the first sub-problem from the classical computer. The quantum computer can then provide a result to the classical computer to re-solve the first problem. The iterative calculation is continued until an end condition is met.