Abstract: A provenance method, system, and non-transitory computer readable medium for a plurality of eidetic systems having logs, include crawling the logs of each node of a plurality of nodes of the eidetic systems to cluster segments across the logs of temporally correlated events into clustered segments and analyzing the correlated segments to interleave an order of processes in the logs and assign a probability to the order of the processes occurring.

Abstract: An artificial intelligence system comprises a computer network server connected to receive and analyze millions of simultaneous text and/or voice messages written by humans to be read and understood by humans. Key, or otherwise important words in sentences are recognized and arrayed. Each such word is contributed to a qualia generator that spawns the word into its possible contexts, themes, or other reasonable ambiguities that can exist at the level of sentences, paragraphs, and missives. A thesaurus-like table is employed to expand each word into a spread of discrete definitions. Several such spreads are used as templates on the others to find petals that exhibit a convergence of meaning. Once the context of a whole missive has been predicted, each paragraph is deconstructed into sub-contexts that are appropriate within the overall theme. Particular contexts identified are then useful to trigger an actionable output.

Abstract: Embodiments for identifying stochastic models representing the individual decision makers in a computing environment by a processor. One or more non-deterministic (stochastic, probabilistic) models may be identified according to a sequence of outcomes from decisions of each of a plurality of decision makers.

Abstract: Record clustering is performed for a collection of records using training rules, training-rule labels, training data created from a sample of pairs of records, a pair-wise classifier, and a clustering algorithm. Record clustering is also performed for a collection of records using prediction rules, prediction-rule labels, a pair-wise classifier, and a clustering algorithm.

Abstract: Systems, methods, and computer program products for selectively customizing, modifying and changing performance parameters, capabilities and behaviors of AI devices within AI device networks by sharing user-selected portions of a knowledge corpus with other AI device networks. Selectively sharing parameters, learned behaviors, capabilities, and features of the knowledge corpus allows for AI devices within a second AI device network to mimic, clone or recreate the performance or behaviors of AI devices operating within the environment of a first AI device network. Users can create portable nodes that travel with the user and upon connecting with the second device network, identify AI devices that can be modified using the portions of the knowledge corpus to recreate selected portions of the first AI device network within the environment of the second AI device network either permanently or for a user-selected amount of time.

Abstract: We present the architecture of a high-performance constraint solver R-Solve that extends the gains made in SAT performance over the past fifteen years on static decision problems to problems that require on-the-fly adaptation, solution space exploration and optimization. R-Solve facilitates collaborative parallel solving and provides an efficient system for unrestricted incremental solving via Smart Repair. R-Solve can address problems in dynamic planning and constrained optimization involving complex logical and arithmetic constraints.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for detecting and analyzing user reactions to messages received. One of the methods includes obtaining reaction data characterizing a reaction of a first user to a communication sent by a second user using a first communication service, wherein the first communication service allows users to react to received communications by selecting from a predetermined set of proprietary reactions that are supported by the first communication service; analyzing the reaction data to generate standardized reaction data that characterizes a sentiment of the reaction of the first user to the communication; mapping the standardized reaction data to one or more proprietary reactions from the predetermined set of proprietary reactions that are supported by the first communication service; and providing, to the first communication service, data identifying the one or more proprietary reactions.

Abstract: One embodiment provides for a computer-readable medium storing instructions that cause one or more processors to perform operations comprising determining a per-layer scale factor to apply to tensor data associated with layers of a neural network model and converting the tensor data to converted tensor data. The tensor data may be converted from a floating point datatype to a second datatype that is an 8-bit datatype. The instructions further cause the one or more processors to generate an output tensor based on the converted tensor data and the per-layer scale factor.

Abstract: Aspects include processors configured to (or include program code that causes a processor to) provide for data classifier devices that extract from structured text business data inputs, via natural language understanding processing, training set data elements (for example, training keywords, training concepts, training entities, and/or training taxonomy classifications, etc.). The aspects identify associations within the structured training business data of each of a plurality of business class categories with respective ones of the extracted training set data elements; and build a logical relationship data classification training knowledge base ontology that connects ones of the business classes to respective associated ones of the extracted training data elements as questions, into a plurality of knowledge base ontology question-business class associations.

Abstract: Data classification extracts from structured text business data inputs, via natural language understanding processing, training set data elements (training keywords, training concepts, training entities, and/or training taxonomy classifications). Embodiments identify associations within the structured training business data of business class categories with respective ones of extracted training set data elements, and build a logical relationship data classification training knowledge base ontology that connects business classes to respective associated ones of extracted training data elements as questions into knowledge base ontology question-business class associations.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training neural networks. In one aspect, a system includes a neural network shrinking engine that is configured to receive a neural network being trained and generate a reduced neural network by a shrinking process. The shrinking process includes training the neural network based on a shrinking engine loss function that includes terms penalizing active neurons of the neural network and removing inactive neurons from the neural network. The system includes a neural network expansion engine that is configured to receive the neural network being trained and generate an expanded neural network by an expansion process including adding new neurons to the neural network and training the neural network based on an expanding engine loss function. The system includes a training subsystem that generates reduced neural networks and expanded neural networks.

Abstract: A method for creating a graph database implemented knowledge mesh is disclosed. The method includes receiving, by a computer system, data from a plurality of different streams and identifying, by the computer system executing at least one machine learning model, a plurality of triples included in the data. The method also comprises filtering, by the computer system, the plurality of triples to identify a relevant subset of triples by applying an ontological filter, applying, by the computer system, a disambiguation routine to the relevant subset of triples to correlate entities included in the relevant subset to other entities in a graph database and determine a degree of confidence with each correlation, and creating, by the computer system, a knowledge mesh using the graph database. The knowledge mesh comprises the relevant subset of triples, each correlation identified by application of the disambiguation routine, and the degree of confidence with each correlation.

Abstract: Techniques are described for classifying seasonal patterns in a time series. In an embodiment, a set of time series data is decomposed to generate a noise signal and a dense signal, where the noise signal includes a plurality of sparse features from the set of time series data and the dense signal includes a plurality of dense features from the set of time series data. A set of one or more sparse features from the noise signal is selected for retention. After selecting the sparse features, a modified set of time series data is generated by combining the set of one or more sparse features with a set of one or more dense features from the plurality of dense features. At least one seasonal pattern is identified from the modified set of time series data. A summary for the seasonal pattern may then be generated and stored.

Abstract: A system for training a neural network. A switch is linked to feature detectors in at least some of the layers of the neural network. For each training case, the switch randomly selectively disables each of the feature detectors in accordance with a preconfigured probability. The weights from each training case are then normalized for applying the neural network to test data.

Abstract: A method is disclosed. The method includes receiving, from a user computer that is a party to a transaction, information that can be used to identify a transaction between the user computer and a resource provider computer. The method further includes determining one or more attributes. The method additionally includes presenting a first question based on the one or more attributes. The method also includes receiving a response to the first question, presenting a second question based on the received response, and receiving a response to the second question. The method further includes storing the received responses in a data storage element, wherein the data storage element is accessible by an authorizing entity computer.

Abstract: Devices and techniques are generally described for generating confidence scores for boosting-based tree machine learning models. In various examples, a first record comprising a plurality of input variables may be received. In another example, a boosting-based tree machine learning model may generate, for the first record, a base model score. In various examples, the base model score may be generated based on the plurality of input variables and the base model score may represent a likelihood that the first record is associated with a first class. In some examples, a score confidence machine learning model may generate a confidence score for the first record. The confidence score may indicate a confidence in the base model score. In various examples, the first record may be processed based at least in part on the confidence score.

Abstract: Systems and methods for embodiments of a graph based artificial intelligence systems for identity management are disclosed. Embodiments of the identity management systems disclosed herein may utilize a network graph approach to analyzing identities, roles, entitlements or other identity management artifacts of a distributed networked enterprise computing environment. Specifically, embodiments of an artificial intelligence based identity management systems may perform predictive modeling for entitlement diffusion or role evolution or other aspects of identity management artifact using network identity graphs.

Abstract: A dynamic prediction model establishment method, an electric device and a user interface are provided. The dynamic prediction model establishment method includes the following steps. An integration model is established by a processing device according to at least one auxiliary data set. The integration model is modified as a dynamic prediction model by the processing device according to a target data set. A sampling point recommendation information is provided by the processing device according to an error degree or an uncertainty degree between the at least one auxiliary data set and the target data set.

Abstract: A computer implemented method, computer program product and system for generating a Bayesian network. A dataset comprising multiple instances of multiple variables is received. A target variable from the received dataset is selected. Multiple parent sets of variables for the target variable are determined, such that, for each parent set of variables, the target variable is functionally dependent on the respective parent set of variables. For multiple variables of the received dataset, the selecting of a new target variable from the received dataset and determining multiple parent sets of variables for the new target variable is repeated. A Bayesian network (includes a directed acyclic graph of nodes and edges) is then generated for the variables such that one or more of the determined parent sets of variables for the target variables are inserted into the graph and edges from the graph are removed to ensure that the graph is acyclic.

Abstract: The present disclosure provides systems and methods for communication efficient distributed mean estimation. In particular, aspects of the present disclosure can be implemented by a system in which a number of vectors reside on a number of different clients, and a centralized server device seeks to estimate the mean of such vectors. According to one aspect of the present disclosure, a client computing device can rotate a vector by a random rotation matrix and then subsequently perform probabilistic quantization on the rotated vector. According to another aspect of the present disclosure, subsequent to quantization but prior to transmission, the client computing can encode the quantized vector according to a variable length coding scheme (e.g., by computing variable length codes).