Abstract: Techniques are provided for evolutionary computer-based optimization and artificial intelligence systems, and include receiving first and second candidate executable code (with ploidy of at least two and one, respectively) each selected at least in part based on a fitness score. The first candidate executable code and the second candidate executable code are combined to produce resultant executable code of the desired ploidy. A fitness score is determined for the resultant executable code, and a determination is made whether the resultant executable code will be used as a future candidate executable code based at least in part on the third fitness score. If an exit condition is met, then the resultant executable code is used as evolved executable code.

Abstract: A system and method for evolving a deep neural network structure that solves a provided problem includes: a memory storing a candidate supermodule genome database having a pool of candidate supermodules having values for hyperparameters for identifying a plurality of neural network modules in the candidate supermodule and further storing fixed multitask neural networks; a training module that assembles and trains N enhanced fixed multitask neural networks and trains each enhanced fixed multitask neural network using training data; an evaluation module that evaluates a performance of each enhanced fixed multitask neural network using validation data; a competition module that discards supermodules in accordance with assigned fitness values and saves others in an elitist pool; an evolution module that evolves the supermodules in the elitist pool; and a solution harvesting module providing for deployment of a selected one of the enhanced fixed multitask neural networks, instantiated with supermodules selected fr

Abstract: A method, computer program product, and system perform computations using a sparse convolutional neural network accelerator. Compressed-sparse data is received for input to a processing element, wherein the compressed-sparse data encodes non-zero elements and corresponding multi-dimensional positions. The non-zero elements are processed in parallel by the processing element to produce a plurality of result values. The corresponding multi-dimensional positions are processed in parallel by the processing element to produce destination addresses for each result value in the plurality of result values. Each result value is transmitted to a destination accumulator associated with the destination address for the result value.

Abstract: A device may receive an instruction to automatically install a program using a click area prediction model. The click area prediction model may be associated with predicting a click area of a user interface that, when selected, causes a program installation procedure to proceed. The device may identify an installation user interface associated with installing the program. The device may determine a group of regions included in the installation user interface. The device may identify sets of features associated with the group of regions. The device may determine, based on the sets of features and the click area prediction model, a group of scores associated with the group of regions. The device may identify a particular region as a predicted click area based on the group of scores. The device may select the predicted click area to attempt to cause the program installation procedure to proceed.

Abstract: Described herein is a framework for outdoor localization. In accordance with one aspect of the framework, a set of hotspot labels are received from one or more user devices connected to an outdoor wireless local area network. Manifold learning may be performed based on the set of hotspot labels to construct one or more manifolds. Using the one or more constructed manifolds, the framework may then estimate a location of a particular user device associated with a query record received from during an online location query.

Abstract: A multiplexed neural core circuit according to one embodiment comprises, for an integer multiplexing factor T that is greater than zero, T sets of electronic neurons, T sets of electronic axons, where each set of the T sets of electronic axons corresponds to one of the T sets of electronic neurons, and a synaptic interconnection network comprising a plurality of electronic synapses that each interconnect a single electronic axon to a single electronic neuron, where the interconnection network interconnects each set of the T sets of electronic axons to its corresponding set of electronic neurons.

Abstract: Technology is directed to text message based concierge services (“the technology”). A user interacts with a concierge service (CS) via text messages to obtain a specific concierge service. For example, the user can send a text message to the CS, e.g., to a contact number provided by the CS, requesting for a recommendation of a restaurant, and the CS can respond by sending the recommendation as a text message. The CS determines a context of the request and generates recommendations that are personalized to the user and is relevant to the context. The CS can use various techniques, e.g., artificial intelligence, machine learning, natural language processing, to determine a context of the request and generate the recommendations accordingly. The CS can also receive additional information from a person associated with the CS, such as a concierge, to further customize or personalize the recommendations to the user.

Abstract: A computer-implemented method for determining the volume of activation of neural tissue. In one embodiment, the method uses one or more parametric equations that define a volume of activation, wherein the parameters for the one or more parametric equations are given as a function of an input vector that includes stimulation parameters. After receiving input data that includes values for the stimulation parameters and defining the input vector using the input data, the input vector is applied to the function to obtain the parameters for the one or more parametric equations. The parametric equation is solved to obtain a calculated volume of activation.

Abstract: A system continuously estimating the state of a dynamical system and classifying signals comprising a computer processor and a computer readable medium having computer executable instructions for providing: a module estimating of the state of a dynamical system assumed to be generated by a Dynamic Movement Primitive; a module classifying signals through inspecting dynamical system state estimates; and a coupling between the two modules such that classifications reset the dynamical system state estimate.

Abstract: Analyzing a set of policies. A goal comprising a particular outcome is received. An analysis object comprising a data structure maintaining information needed to perform an analysis of the goal is defined. The analysis object is configured to limit a number of calculations needed to achieve the goal. Each member of a set of expressions found in the set of policies has an output. The output is the same for each expression. One of the set of expressions is solved. The solved output is cached in the analysis object such that the solved output is associated with each member of the set of expressions. The analysis object is processed to create a set of values that achieves the goal. Processing includes referencing the cache to retrieve the solved output each time a member of the set of expressions is to be solved during processing of the analysis object.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a reinforcement learning system. The method includes: training an action selection policy neural network, and during the training of the action selection neural network, training one or more auxiliary control neural networks and a reward prediction neural network. Each of the auxiliary control neural networks is configured to receive a respective intermediate output generated by the action selection policy neural network and generate a policy output for a corresponding auxiliary control task. The reward prediction neural network is configured to receive one or more intermediate outputs generated by the action selection policy neural network and generate a corresponding predicted reward.

Abstract: A resistive memory cell is connected in circuitry which has a first input terminal for applying neuron input signals including a read portion and a write portion. The circuitry includes a read circuit producing a read signal dependent on resistance of the memory cell, and an output terminal providing a neuron output signal, dependent on the read signal in a first state of the memory cell. The circuitry also includes a storage circuit storing a measurement signal dependent on the read signal, and a switch set operable to supply the read signal to the storage circuit during application of the read portion of each neuron input signal to the memory cell, and, after application of the read portion, to apply the measurement signal in the apparatus to enable resetting of the memory cell to a second state.

Abstract: A heterogeneous log pattern editing recommendation system and computer-implemented method are provided. The system has a processor configured to identify, from heterogeneous logs, patterns including variable fields and constant fields. The processor is also configured to extract a category feature, a cardinality feature, and a before-after n-gram feature by tokenizing the variable fields in the identified patterns. The processor is additionally configured to generate target similarity scores between target fields to be potentially edited and other fields from among the variable fields in the heterogeneous logs using pattern editing operations based on the extracted category feature, the extracted cardinality feature, and the extracted before-after n-gram feature. The processor is further configured to recommend, to a user, log pattern edits for at least one of the target fields based on the target similarity scores between the target fields in the heterogeneous logs.

Abstract: Systems, methods, and computer media for machine learning through a symbolic, parallelized stochastic gradient descent (SGD) analysis are provided. An initial data portion analyzer can be configured to perform, using a first processor, SGD analysis on an initial portion of a training dataset. Values for output model weights for the initial portion are initialized to concrete values. Local model builders can be configured to perform, using an additional processor for each local model builder, symbolic SGD analysis on an additional portion of the training dataset. The symbolic SGD analysis uses a symbolic representation as an initial state for output model weights for the corresponding portions of the training dataset. The symbolic representation allows the SGD analysis and symbolic SGD analysis to be performed in parallel. A global model builder can be configured to combine outputs of the local model builders and the initial data portion analyzer into a global model.

Abstract: The technology disclosed introduces a concept of training a neural network to create an embedding space. The neural network is trained by providing a set of K+2 training documents, each training document being represented by a training vector x, the set including a target document represented by a vector xt, a favored document represented by a vector xs, and K>1 unfavored documents represented by vectors xiu, each of the vectors including input vector elements, passing the vector representing each document set through the neural network to derive an output vectors yt, ys and yiu, each output vector including output vector elements, the neural network including adjustable parameters which dictate an amount of influence imposed on each input vector element to derive each output vector element, adjusting the parameters of the neural network to reduce a loss, which is an average over all of the output vectors yiu of [D(yt,ys)?D(yt, yiu)].

Abstract: Generally discussed herein are devices, systems, and methods for machine-learning. A method may include projecting an input feature vector of a first dimensional space into a second dimensional space to create a lower dimensional feature vector, the second dimensional space smaller than the first dimensional space, determining a first prediction vector for an internal node of the tree, determining whether to pass the first prediction vector to a first child or a second child of the internal node based on a sparse vector and the lower dimensional feature vector, determining a second prediction vector at a leaf node of the tree, and determining an overall prediction by combining the first and second prediction vectors.

Abstract: Regularization of neural networks. Neural networks can be regularized by obtaining an original neural network having a plurality of first-in-first-out (FIFO) queues, each FIFO queue located between a pair of nodes among a plurality of nodes of the original neural network, generating at least one modified neural network, the modified neural network being equivalent to the original neural network with a modified length of at least one FIFO queue, evaluating each neural network among the original neural network and the at least one modified neural network, and determining which neural network among the original neural network and the at least one modified neural network is most accurate, based on the evaluation.

Abstract: An endless loop in an inference engine processing a set of rules according to facts stored in a database may be detected by: (a) analyzing the set of rules to identify a subset of the set of rules comprising rules that are triggered by an updated fact; (b) executing the rules of the subset; (c) updating at least one of the facts based on the execution of the rules; (d) storing an identifier for each executed rule of the subset; (e) associating each stored identifier with a subset number for the subset in a consecutive sequence of subset numbers for executed subsets; and repeating steps (a)-(e) until the identifier for a rule, to be executed, is associated with an excessive number of subset numbers that are equally offset from each other in the sequence of subset numbers for executed subsets.

Abstract: A method and system for assigning a multimedia content element to a user. The method includes generating at least one signature to the multimedia content element; determining, based on the generated at least one signature, whether the multimedia content element exists in a database; and assigning a unique identifier of the user to the multimedia content element, when it is determined that the multimedia content element does not exist in the database.

Abstract: An individual neuron circuit calculates a first value based on a sum of products each obtained by multiplying one of weight values, each representing connection or disconnection between a corresponding neuron circuit and one of the other neuron circuits, by a corresponding one of output signals of the other neuron circuits and outputs 0 or 1, based on a result of comparison between a second value obtained by adding a noise value to the first value and a threshold. An arbitration circuit allows, when first output signals of first neuron circuits interconnected among the neuron circuits simultaneously change based on the weight values, updating of only one of the first output signals of the first neuron circuits and allows, when second output signals of second neuron circuits not interconnected simultaneously change, updating of the second output signals.