Abstract: A method for selecting an action, includes reading, into a memory, a Partially Observed Markov Decision Process (POMDP) model, the POMDP model having top-k action IDs for each belief state, the top-k action IDs maximizing expected long-term cumulative rewards in each time-step, and k being an integer of two or more, in the execution-time process of the POMDP model, detecting a situation where an action identified by the best action ID among the top-k action IDs for a current belief state is unable to be selected due to a constraint, and selecting and executing an action identified by the second best action ID among the top-k action IDs for the current belief state in response to a detection of the situation. The top-k action IDs may be top-k alpha vectors, each of the top-k alpha vectors having an associated action, or identifiers of top-k actions associated with alpha vectors.

Abstract: A non-transitory computer-readable storage medium storing a program that causes a computer to execute a process, the process includes learning a learning model using a first training data group obtained by excluding a test data group from a plurality of training data items; calculating prediction accuracy of the learning model using the test data group; and when the prediction accuracy satisfies the predetermined requirement, learning an error prediction model for determining whether an error of a value predicted by the learning model satisfies a predetermined requirement, by using a second training data group obtained by excluding the test data group and the first training data group from the plurality of training data items.

Abstract: Embodiments can provide a computer implemented method for harvesting training data for a training set for use by a system capable of answering questions, the system comprising a processor and a memory comprising instructions executed by the processor, the method comprising receiving, from a user, an input question; processing the input question and returning, to the user, a result set comprising one or more ranked hypotheses and one or more ranked evidence passages corresponding to the one or more ranked hypotheses; receiving, from the user, an indication that one of the one or more ranked hypotheses is to be designated a watched hypothesis; adding the input question and the watched hypothesis to a to-be-vetted question/answer (QA) pair set comprising one or more to-be-vetted QA pairs; vetting each of the one or more to-be-vetted QA pairs in the to-be-vetted QA pair set through a first-pass automatic vetting procedure; if a vetted QA pair passes the first-pass automatic vetting procedure, adding the vetted QA

Abstract: A machine learning hardware accelerator architecture and associated techniques are disclosed. The architecture features multiple memory banks of very wide SRAM that may be concurrently accessed by a large number of parallel operational units. Each operational unit supports an instruction set specific to machine learning, including optimizations for performing tensor operations and convolutions. Optimized addressing, an optimized shift reader and variations on a multicast network that permutes and copies data and associates with an operational unit that support those operations are also disclosed.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing inputs using recurrent neural networks. One of the systems includes a main recurrent neural network comprising one or more recurrent neural network layers and a respective hyper recurrent neural network corresponding to each of the one or more recurrent neural network layers, wherein each hyper recurrent neural network is configured to, at each of a plurality of time steps: process the layer input at the time step to the corresponding recurrent neural network layer, the current layer hidden state of the corresponding recurrent neural network layer, and a current hypernetwork hidden state of the hyper recurrent neural network to generate an updated hypernetwork hidden state.

Abstract: A device, method, and non-transitory computer readable storage medium for labelling motion data are provided. The device receives several motion signals, wherein each motion signal includes a motion time message and a motion data group. A motion script includes a plurality of preset motion messages, wherein each preset motion message includes a preset time message and a preset motion. The device performs the following steps for each preset time message: determining a first subset of the motion signals by comparing the motion time messages with the preset time message, calculating a similarity between the motion data group of each motion signal in the first subset and a reference model, determining a second subset of the first subset based on the first similarities, and labelling the motion data group of each motion signal included in the second subset as corresponding to the preset motion corresponding to the preset time message.

Abstract: A system and method for tuning hyperparameters and training a model includes implementing a hyperparameter tuning service that tunes hyperparameters of a model that includes receiving, via an API, a tuning request that includes: (i) a first part comprising tuning parameters for generating tuned hyperparameter values for hyperparameters of the model; and (ii) a second part comprising model training control parameters for monitoring and controlling a training of the model, wherein the model training control parameters include criteria for generating instructions for curtailing a training run of the model; monitoring the training run for training the model based on the second part of the tuning request, wherein the monitoring of the training run includes periodically collecting training run data; and computing an advanced training curtailment instruction based on the training run data that automatically curtails the training run prior to a predefined maximum training schedule of the training run.

Abstract: In an example embodiment, a solution that creates a questionnaire mapping record for questions in a computerized document is utilized to map questions in the computerized document to normalized questions. Where necessary, normalized questions can be automatically created and included in the questionnaire mapping record. Handing strategy rules may also be automatically created for the normalized question, with the handling strategy rules defining how data may be automatically retrieved and used to prepopulate answers to the questions in the computerized document.

Abstract: Cascaded, boosted predictive models trained using distinct sets of exogenous and endogenous features are configured to predict component of performance ratings of entities. From the distinct predicted components, the second entity's rating factor can be determined. A second entity's rating factor represents the specific contribution a second entity makes to his average performance rating, as distinct from the rating that an arbitrary or hypothetical second entity would obtain.

Abstract: Disclosed herein is a system, which comprises a plurality of processing units. Each of the processing units comprises a first oscillator, a second oscillator, and a counter. Each of the processing units is configured to receive a first input and a second input and to send an output as a function of the first input and the second input. The function has a plurality of parameters. Each of the processing units is configured to receive and send values of the parameters. The system can be used together with a microprocessor to perform parallel computing.

Abstract: Cardinal sine function used as an activation function for a hierarchical classifier. Application of a sine function, or a cardinal sine function, for hierarchical classification of a subject within subject matter domains and sub-domains. Hierarchical classification or multi-level classification is improved through use of the cardinal sine function or even standard sine function. Some embodiments of the present invention focus on the usage of cardinal sine function as activation function and how to apply this cardinal sine function for hierarchical classification of a subject. Some embodiments include a technique by which hierarchical classification or multi-level classification can benefit from application of a cardinal sine function.

Abstract: Systems and methods for transforming legacy models and transforming a model into a neural network model are disclosed. In an embodiment, a method may include receiving input data comprising an input model, an input dataset, and an input command. The method may include applying the input model to the input dataset to generate model output and storing model output and at least one of input model features or a map of the input model. The method may include generating a candidate neural network models with parameters. The method may include tuning the candidate neural network models to the input model. The method may include receiving model output from the candidate neural network models and selecting a neural network model from the candidate neural network models based on the candidate model output and the model selection criteria. In some aspects, the method may include returning the selected neural network model.

Abstract: The description relates to predicting terms based on text inputted by a user. One example includes a computing device comprising a processor configured to send, over a communications network, the text to a remote prediction engine. The processor is configured to send the text to a local prediction engine stored at the computing device, and to monitor for a local predicted term from the local prediction engine and a remote predicted term from the remote prediction engine, in response to the sent text. The computing device includes a user interface configured to present a final predicted term to the user such that the user is able to select the final term. The processor is configured to form the final predicted term using either the remote predicted term or the local predicted term on the basis of a time interval running from the time at which the user input the text.

Abstract: Enterprise Hypothesis Orchestration provides users an intuitive system for building an inquiry model that thereafter creates and evaluates each of a plurality of hypotheses as it continuously searches for evidence to formulate, score, and resolve each hypothesis. The Enterprise Hypothesis Orchestration system moreover continuously deals with the uncertainty caused by noisy, missing, inaccurate, and/or contradictory data. The present invention uses abductive reasoning to infer the best explanation or hypothesis for a set of observations. Given an inquiry the Hypothesis Orchestration System identifies relevant data from which to form a plurality of hypotheses. It thereafter collects evidence in support of each hypothesis and crafts a degree of confidence that the hypothesis is true. If a hypothesis is found to lack support an analysis of any missing evidence is conducted to identify and seek which evidence would offer the highest benefit to resolving one or more of the plurality of hypotheses.

Abstract: Systems, devices, and methods of the present invention are related to determining a document classification. For example, a document classification application generates a set of discourse trees, each discourse tree corresponding to a sentence of a document and including a rhetorical relationship that relates two elementary discourse units. The document classification application creates one or more communicative discourse trees from the discourse trees by matching each elementary discourse unit in a discourse tree that has a verb to a verb signature. The document classification application combines the first communicative discourse tree and the second communicative discourse tree into a parse thicket and applies a classification model to the parse thicket in order to determine whether the document is public or private.

Abstract: A system rapidly produces training cases for machine based learning by automatically creating training cases from a database of historical data. The system determines a plurality of attributes relevant to each of the training cases. The system identifies a first attribute of the plurality of attributes as an issue, and a second attribute of the plurality attributes as a response to the issue. The system identifies a plurality of cohort members from the database of historical data, where each cohort member comprises cohort member attributes that match a subset of the plurality of attributes. The system analyzes the cohort member attributes of each of the plurality of cohort members to identify the most frequent responses to the issue. The system creates the training cases where each training case comprises the issue and the most frequent responses. The system then trains a machine based learning system using the training cases.

Abstract: A property vector representing extractable measurable properties, such as musical properties, of a file is mapped to semantic properties for the file. This is achieved by using artificial neural networks “ANNs” in which weights and biases are trained to align a distance dissimilarity measure in property space for pairwise comparative files back towards a corresponding semantic distance dissimilarity measure in semantic space for those same files. The result is that, once optimised, the ANNs can process any file, parsed with those properties, to identify other files sharing common traits reflective of emotional-perception, thereby rendering a more liable and true-to-life result of similarity/dissimilarity. This contrasts with simply training a neural network to consider extractable measurable properties that, in isolation, do not provide a reliable contextual relationship into the real-world.

Abstract: The present disclosure is directed to filtering co-occurrence data. In one embodiment, a machine learning model can be trained. An output of an intermediate structure of the machine learning model (e.g., an output of an internal layer of a neural network) can be used as a representation of an event. Similarities between representations of events can be determined and used to generate, augment, or modify co-occurrence data.

Abstract: A computer implemented method identifies guidelines through use of a neural network by a remote guideline server. A client computer transmits instructions to the remote guideline server to retrieve and evaluate multiple candidate guidelines. The remote guideline server utilizes a neural network to identify a string of terms found in each of the multiple candidate guidelines that match one or more strings of terms from a model guideline; to identify a semantic concept of each of the multiple candidate guidelines that matches one or more semantic concepts from the model guideline; and to identify a structural pattern of each of the multiple candidate guidelines that matches one or more structural patterns of the model guideline. The candidate guidelines that match the model guideline are then sent from the remote guideline server to the client computer.

Abstract: An implementation of neural networks on silicon for the processing of various signals comprises multidimensional signals such as images. The efficient implementation on silicon of a complete processing chain for the signal via the approach using neural networks is provided. The circuit comprises at least: a series of neuro-blocks grouped together in branches composed of a group of neuro-blocks and a broadcasting bus, the neuro-blocks connected to the broadcasting bus; a routing unit connected to the broadcasting bus of the branches, carrying out the routing and broadcasting of data to and from the branches; a transformation module connected to the routing unit via an internal bus and designed to be connected at the input of the circuit to an external databus, the module carrying out the transformation of input data into serial coded data. The processing operations internal to the circuit are carried out according to a serial communications protocol.