Abstract: Disclosed systems and methods generate user-session representation vectors from data generated by user interactions with online services. A transformation application executing on a computing device receives interaction data, which is generated by user devices interacting with an online service. The transformation application separates the interaction data into session datasets. The transformation involves normalizing the session datasets by modifying the rows within each session dataset by removing event identifiers and time stamps. The application transforms each normalized session dataset into a respective user-session representation vector. The application outputs the user-session representation vectors.

Abstract: An integrated circuit includes a data storage circuit, a weight storage circuit, and an operation circuit. The data storage circuit stores, according to a preset control instruction, gray-scale value data of an image in a first preset time period; the weight storage circuit stores a data weight corresponding to the gray-scale value data in the first preset time period; the operation circuit carries out an operation on the gray-scale value data and the data weight in the first preset time period and outputs first data; the data storage circuit further stores the first data according to the preset control instruction in a second preset time period; the weight storage circuit further stores the data weight corresponding to the first data in the second preset time period; the operation circuit carries out an operation on the first data and the data weight in the second time period and outputs second data.

Abstract: A method for learning a pedestrian detector to be used for robust surveillance or military purposes based on image analysis is provided for a solution to a lack of labeled images and for a reduction of annotation costs. The method can be also performed by using generative adversarial networks (GANs). The method includes steps of: a learning device generating an image patch by cropping each of regions on a training image, and instructing an adversarial style transformer to generate a transformed image patch by converting each of pedestrians into transformed pedestrians capable of impeding a detection; and generating a transformed training image by replacing each of the regions with the transformed image patch, instructing the pedestrian detector to detecting the transformed pedestrians, and learning parameters of the pedestrian detector to minimize losses. This learning, as a self-evolving system, is robust to adversarial patterns by generating training data including hard examples.

Abstract: A neuromorphic device includes a synapse. The synapse includes a first electrode, a second electrode spaced apart from the first electrode, an oxygen-containing layer disposed between the first electrode and the second electrode, the oxygen-containing layer including oxygen ions, and a stack structure disposed between the oxygen-containing layer and the second electrode, the stack structure including a plurality of reactive metal layers alternately arranged with a plurality of oxygen diffusion-retarding layers. The plurality of reactive metal layers are capable of reacting with oxygen ions of the oxygen-containing layer. The plurality of oxygen diffusion-retarding layers interfere with a movement of the oxygen ions from the oxygen-containing layer to the plurality of reactive metal layers.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a neural network used to select actions performed by a reinforcement learning agent interacting with an environment. In one aspect, a method includes maintaining a replay memory, where the replay memory stores pieces of experience data generated as a result of the reinforcement learning agent interacting with the environment. Each piece of experience data is associated with a respective expected learning progress measure that is a measure of an expected amount of progress made in the training of the neural network if the neural network is trained on the piece of experience data. The method further includes selecting a piece of experience data from the replay memory by prioritizing for selection pieces of experience data having relatively higher expected learning progress measures and training the neural network on the selected piece of experience data.

Abstract: A method and a system are provided for correlation detection in multiple spatio-temporal datasets for event sensing in a geographical area. The method includes extracting datasets, comprising information about one or more events, from one or more data sources. The method further includes identifying a primary data source and secondary data sources from the one or more data sources. The method further includes extracting primary features from the datasets associated with the primary data source and secondary features from the datasets associated with the secondary data sources. The primary features are categorized into one or more categories. The method further includes training classifiers based on the primary features and/or the one or more categories. The method further includes detecting a correlation among the information associated with the one or more events based on a category transfer distribution from the primary data source to the secondary data sources.

Abstract: A source system initializes, using an initialization seed, a first parameter vector representing weights of a neural network. The source system determines a second parameter vector by performing a sequence of mutations on the first parameter vector, the mutations each being based on a perturbation seed. The source system generates, and stores to memory, an encoded representation of the second parameter vector that comprises the initialization seed and a sequence of perturbation seeds corresponding to the sequence of mutations. The source system transmits the data structure to a target system, which processes a neural network based on the data structure.

Abstract: Methods and systems for time-series transient analysis of data are disclosed herein. A method includes receiving time-series data, generating a training data set including randomized data points, generating randomized data point combinations using a set of the randomized data points that are within a time window, computing distance values based on the randomized data point combinations, generating a classifier based on a plurality of computed distance values, and determining, using the classifier, a probability that new time-series data generated during a new execution of the process matches the time-series data. A system for performing the method is also disclosed.

Abstract: A mechanism is provided in a computing device configured with instructions executing on a processor of the computing device to implement a ground truth generation system for quality assessment scoring of articles in a corpus. The ground truth generation system receives recommendations of a set of recommended articles from subject matter experts. The ground truth generation system identifies a set of non-recommended articles. A topic clustering component within the ground truth generation system performs topic clustering on a combination of the set of recommended articles and the set of non-recommended articles to form a set of topic clusters containing recommended articles and non-recommended articles. The ground truth generation system identifies a first number of recommended articles and a second number of non-recommended articles in each of the set of topic clusters to form a quality assessment training set.

Abstract: A causal relationship is represented with a data structure including target identification information for identifying a target, index identification information of each of indexes to be used for quantitatively describing an event that occurs to the target, and causal relationship information for each pair of two indexes selected from the indexes. The causal relationship information includes direction information representing which one of the two indexes is a cause index and which one is an effect index, strength information representing a causal strength between the cause index and the effect index, time information representing a delay time for propagation of an influence of the cause index to the effect index, and correlation information representing a direction of change in the effect index with respect to an increase or decrease in the cause index.

Abstract: The present disclosure relates to the electronic document review field and, more particularly, to various apparatuses and methods of implementing batch-mode active learning for technology-assisted review (TAR) of documents (e.g., legal documents).

Abstract: The present disclosure relates to the electronic document review field and, more particularly, to various apparatuses and methods of implementing batch-mode active learning for technology-assisted review (TAR) of documents (e.g., legal documents).

Abstract: Certain aspects involve optimizing neural networks or other models for assessing risks and generating explanatory data regarding predictor variables used in the model. In one example, a system identifies predictor variables compliant with certain monotonicity constraints. The system generates a neural network for determining a relationship between each predictor variable and a risk indicator. The system performs a factor analysis on the predictor variables to determine common factors. The system iteratively adjusts the neural network so that (i) a monotonic relationship exists between each common factor and the risk indicator and (ii) a respective variance inflation factor for each common factor is sufficiently low. Each variance inflation factor indicates multicollinearity among a subset of the predictor variables corresponding to a common factor.

Abstract: A policy violation detection computer-implemented method, system, and computer program product, includes extracting a policy activity from a policy, the policy activity including an actor in the policy, an object of the policy, an action of the policy, and policy scope metadata, capturing a transaction by a user including metadata of the transaction, translating the transaction by the user into an actor in the transaction, an action of the transaction, and an object of the transaction, and alerting the user of a policy violation by navigating a knowledge graph is-a hierarchy to relate the actor in the transaction to the actor in the policy, the object of the transaction to an object of the policy, and the action of the transaction to an action of the policy activity.