Abstract: Disclosed are techniques for generating a personalized flow for a software delivery model. These techniques identify user information expressed in natural language for a specific user. One or more user clusters may be determined for the specific user based on at least one user vector representation of a form of the user information. One or more personalized information clusters may be identified for a user cluster of the one or more user clusters based at least in part on the at least one user vector representation. A personalized software application flow may be generated and presented to the specific user by using at least the one or more personalized information clusters for the specific user.

Abstract: Disclosed in some examples are methods, systems, machine-readable media, and devices which operate a neural network defined by user code. A method includes identifying, operations from user code that are integral in operating the neural network, combining a subset of the identified operations into a single processing sequence to be transmitted to an array of hardware processors, performing operations that are not integral in operation of the neural network in a separate thread of execution from the operations that are integral in operating the neural network; and mapping results to the combined operations that were included in the single processing sequence.

Abstract: At a network-accessible artificial intelligence service for time series predictions, a recurrent neural network model is trained using a plurality of time series of demand observations to generate demand forecasts for various items. A probabilistic demand forecast is generated for a target item using multiple executions of the trained model. Within the training set used for the model, the count of demand observations of the target item may differ from the count of demand observations of other items. A representation of the probabilistic demand forecast may be provided via a programmatic interface.

Abstract: In one embodiment, a processor is to store a membrane potential of a neural unit of a neural network; and calculate, at a particular time-step of the neural network, a change to the membrane potential of the neural unit occurring over multiple time-steps that have elapsed since the last time-step at which the membrane potential was updated, wherein each of the multiple time-steps that have elapsed since the last time-step is associated with at least one input to the neural unit that affects the membrane potential of the neural unit.

Abstract: A regression model is generated to map observation records of a first dimensionality to a second dimensionality. Using a set of transformed records obtained from the first regression model, a Gaussian mixture model of the distribution of observation records of the second dimensionality is trained. Using a Gaussian distribution obtained from the Gaussian mixture model, a recommended modification of a proposed training set of a classifier is obtained.

Abstract: A device includes a state machine. The state machine includes a plurality of blocks, where each of the blocks includes a plurality of rows. Each of these rows includes a plurality of programmable elements. Furthermore, each of the programmable elements are configured to analyze at least a portion of a data stream and to selectively output a result of the analysis. Each of the plurality of blocks also has corresponding block activation logic configured to dynamically power-up the block.

Abstract: Methods and apparatus are provided for taking snapshots to train prediction models and improve workflow execution.

Abstract: Some aspects are directed to systems for providing access to data. An example includes a computer system implementing a data access framework for providing data to one or more predictive models. The system is configured to receive at least one asset definition comprising an asset identifier for at least one asset associated with at least one data source, receive at least one data lens definition comprising a data lens identifier and a logical model identifier, the logical model identifier identifying at least one portion of a logical system model, access a first datastore to retrieve asset information extracted from the at least one data source, determine, using the logical model identifier, a portion of the logical system model associated with the retrieved asset information, format the retrieved asset information for storage in a second datastore corresponding to the logical system model, and store the formatted retrieved asset information in the second datastore.

Abstract: A method of inverse reinforcement learning for estimating cost and value functions of behaviors of a subject includes acquiring data representing changes in state variables that define the behaviors of the subject; applying a modified Bellman equation given by Eq. (1) to the acquired data: q(x)+gV(y)?V(x)=?ln{pi(y|x))/(p(y|x)} (1) where q(x) and V(x) denote a cost function and a value function, respectively, at state x, g represents a discount factor, and p(y|x) and pi(y|x) denote state transition probabilities before and after learning, respectively; estimating a density ratio pi(y|x)/p(y|x) in Eq. (1); estimating q(x) and V(x) in Eq. (1) using the least square method in accordance with the estimated density ratio pi(y|x)/p(y|x), and outputting the estimated q(x) and V(x).

Abstract: A method of inverse reinforcement learning for estimating reward and value functions of behaviors of a subject includes: acquiring data representing changes in state variables that define the behaviors of the subject; applying a modified Bellman equation given by Eq. (1) to the acquired data: r ? ( x ) + ? ? ? V ? ( y ) - V ? ( x ) = ? ln ? ? ? ? ( y | x ) b ? ( y | x ) , ? ( 1 ) = ? ln ? ? ? ? ( x , y ) b ? ( x , y ) - ln ? ? ? ? ( x ) b ? ( x ) ,                                                ? ( 2 ) where r(x) and V(x) denote a reward function and a value function, respectively, at state x, and ? represents a discount factor, and b(y|x) and ?(y|x) denote state transition probabilities before and after learning, respectively; estimating a logarithm of the density ratio ?(x)/b(x) in Eq. (2); estimating r(x) and V(x) in Eq.

Abstract: Systems and methods of leveraging machine learning to predict user generated content are provided. For instance, first entity data associated with an entity can be received. The first entity data can include user specified data associated with an attribute of the entity. The first entity data can be input into a machine-learned content prediction model. Inferred entity data can be received as output of the machine-learned content prediction model. The inferred entity data can include inferred data descriptive of the attribute of the entity.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training an action selection policy neural network. In one aspect, a method comprises: obtaining an expert observation; processing the expert observation using a generative neural network system to generate a given observation-given action pair, wherein the generative neural network system has been trained to be more likely to generate a particular observation-particular action pair if performing the particular action in response to the particular observation is more likely to result in the environment later reaching the state characterized by a target observation; processing the given observation using the action selection policy neural network to generate a given action score for the given action; and adjusting the current values of the action selection policy neural network parameters to increase the given action score for the given action.

Abstract: Techniques to response to respond to user requests using natural-language machine learning based on branching example conversations are described.

Abstract: A computing system trains a clustering model. A responsibility parameter vector is initialized for each observation vector and includes a probability value of a cluster membership. The observation vectors include a plurality of classified observation vectors and a plurality of unclassified observation vectors. (A) Beta distribution parameter values are computed for each cluster. (B) Parameter values are computed for a normal-Wishart distribution for each cluster. (C) Each responsibility parameter vector is updated using the beta distribution parameter values, the parameter values, and a respective observation vector. (D) A convergence parameter value is computed. (E) (A) to (D) are repeated until the computed convergence parameter value indicates the responsibility parameter vector defined for each observation vector of the plurality of unclassified observation vectors is converged.

Abstract: A deep learning-based autonomous vehicle control system includes: a processor determining an autonomous driving control based on deep learning, correcting an error in determination of the deep learning-based autonomous driving control based on determination of an autonomous driving control based on a predetermined expert rule, and controlling an autonomous vehicle; and a non-transitory computer-readable storage medium storing data for the determination of the deep learning-based autonomous driving control, data for the determination of the expert rule-based autonomous driving control, and information about the error in the determination of the deep learning-based autonomous driving control.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for selecting an action to be performed by a reinforcement learning agent interacting with an environment. A current observation characterizing a current state of the environment is received. For each action in a set of multiple actions that can be performed by the agent to interact with the environment, a probability distribution is determined over possible Q returns for the action-current observation pair. For each action, a measure of central tendency of the possible Q returns with respect to the probability distributions for the action-current observation pair is determined. An action to be performed by the agent in response to the current observation is selected using the measures of central tendency.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for controlling an agent interacting with an environment. One of the methods includes obtaining a representation of an observation; processing the representation using a convolutional long short-term memory (LSTM) neural network comprising a plurality of convolutional LSTM neural network layers; processing an action selection input comprising the final LSTM hidden state output for the time step using an action selection neural network that is configured to receive the action selection input and to process the action selection input to generate an action selection output that defines an action to be performed by the agent at the time step; selecting, from the action selection output, the action to be performed by the agent at the time step in accordance with an action selection policy; and causing the agent to perform the selected action.

Abstract: Techniques for solving a multi-agent plan recognition problem are provided. In one example, a computer-implemented method comprises transforming, by a device operatively coupled to a processor, a problem model and an at least partially ordered sequence of observations into an artificial intelligence planning problem through a transform algorithm. The problem model can comprises a domain description from a plurality of agents and a durative action. Furthermore, at least one of the observations of the at least partially ordered sequence of observations can be a condition that changes over time. The computer-implemented method further comprises determining, by the device, plan information using an artificial intelligence planner on the artificial intelligence planning problem. The computer-implemented method further comprises translating, by the device, the plan information into information indicative of a solution to the artificial intelligence planning problem.

Abstract: A method of continuous decoding of motion for a direct neural interface. The method of decoding estimates a motion variable from an observation variable obtained by a time-frequency transformation of the neural signals. The observation variable is modelled using a HMM model whose hidden states include at least an active state and an idle state. The motion variable is estimated using a Markov mixture of experts where each expert is associated with a state of the model. For a sequence of observation vectors, the probability that the model is in a given state is estimated, and from this a weighting coefficient is deduced for the prediction generated by the expert associated with this state. The motion variable is then estimated by combination of the estimates of the different experts with these weighting coefficients.

Abstract: A learning method for supporting a safer autonomous driving through a fusion of information acquired from images and communications is provided. And the method includes steps of: (a) a learning device instructing a first neural network and a second neural network to generate an image-based feature map and a communication-based feature map by using a circumstance image and circumstance communication information; (b) the learning device instructing a third neural network to apply a third neural network operation to the image-based feature map and the communication-based feature map to generate an integrated feature map; (c) the learning device instructing a fourth neural network to apply a fourth neural network operation to the integrated feature map to generate estimated surrounding motion information; and (d) the learning device instructing a first loss layer to train parameters of the first to the fourth neural networks.