Abstract: Methods and systems described herein for validating machine learning models in federated machine learning model environments. More specifically, the methods and systems relate to unloading training and validation techniques to client devices using newly collected data to improve accuracy of federated machine learning models.

Abstract: A system enables agile model development to speed up innovation by data scientists. Model training and deployment are coordinated and standardized to reduce redundancy. Data is obtained for feature generation and reformatted and de-sensitized for storage. The features are stored in locations available to all models and training modules of a system so data does not need to be adjusted for new models. To generate a machine learning model, the system establishes a cohort for evaluation by the model. A model template and features for use by the model are identified. The selected template and features are used for experimentation and evaluation. Model training artifacts, such as model weights are subsequently recorded in a model store and the model scripts and settings can then be registered in a centralized database where it can be accessed for execution.

Abstract: Methods and systems for learning policies using sparse and underspecified rewards. One of the methods includes training the policy jointly with an auxiliary reward function having a plurality of auxiliary reward parameters, the auxiliary reward function being configured to map, in accordance with the auxiliary reward parameters, trajectory features of at least a trajectory to an auxiliary reward value that indicates how well the trajectory performed a task in response to a context input.

Abstract: Methods and systems are disclosed for performing operations for providing a shared augmented reality unboxing experience. The operations include causing concurrent display of a shared augmented reality experience comprising a shared virtual box that is in a closed state on a plurality of client devices associated with a plurality of users and obtaining a sequence of triggers associated with the shared virtual box. First and second inputs are received respectively from first and second client devices. The operations include determining that the first and second inputs correspond to the sequence of triggers associated with the shared virtual box. The operations include modifying the shared virtual box from being displayed on the plurality of devices in the closed state to being displayed in the open state.

Abstract: Embodiments herein disclose, e.g., a method performed by a control network node in a communications network for handling machine learning (ML) models in the communications network. The control network node determines whether or not to transmit to a network node in the communications network a ML model based on a signature and/or a loss value of the network node, wherein the signature and/or the loss value is related to ML modelling. In case where it is determined to transmit, the control network node transmits the ML model to the network node.

Abstract: This disclosure relates generally to system and method for molecular property prediction. Typically, message-pooling mechanism employed in molecular property prediction using conventional message passing neural networks (MPNN) causes over smoothing of the node embeddings of the molecular graph. The disclosed system utilizes edge conditioned identity mapping convolution neural network for the message passing phase. In message passing phase, the system computes an incoming aggregated message vector for each node of the plurality of nodes of the molecular graph based on encoded message received from neighboring nodes such that encoded message vector is generated by fusing a node information and an connecting edge information of the set of neighboring nodes of the node. The incoming aggregated message vector is utilized for computing updated hidden state vector of each node.

Abstract: A training method of an autoencoder that performs encoding and decoding, for a computer to execute a process includes encoding input data by the autoencoder; obtaining a probability distribution of feature data obtained by encoding the input data; generating first decoded data by decoding the feature data by the autoencoder; adding a noise to the feature data by the autoencoder; generating second decoded data by decoding the feature data to which the noise is added by the autoencoder; and training the autoencoder to train the probability distribution of the feature data so that a first error between the first decoded data and the input data, a second error between the first decoded data and the second decoded data, and an information entropy of the probability distribution are decreased.

Abstract: An exploration method used by an exploration apparatus in multi-agent reinforcement learning to collect training samples during the training process is provided. The exploration method includes calculating the influence of a selected action of each agent on the actions of other agents in a current state, calculating a linear sum of the value of a utility function representing the action value of each agent and the influence on the actions of the other agent calculated for the selected action of each agent, and obtaining a sample to be used for training an action policy of each agent by probabilistically selecting the action in which the linear sum is the maximum, and the random action.

Abstract: A general-purpose approach to solving the core problems of detecting and predicting the actions of invisible actors, and the consequential challenges of intervention and prevention. The operational forecasting system is applied to data gathered from complex systems. The operational forecasting system uses novel early-warning signals that are based on anomalous behaviors of actors/agents that are observed, as they respond to those unobserved actors that are the source of systemic change. The operational forecasting system targets predicting when an event will occur, before it does, based on the anomalous behaviors of observed actors responding to those invisible actors that are creating the perturbation (i.e. the murmuration).

Abstract: A summarization system includes: K embedding modules configured to: receive K blocks of text, respectively, of a document to be summarized; and generate K first representations based on the K blocks of text, respectively, where K is an integer greater than 2; a first propagation module configured to generate second representations based on the K first representations; a second propagation module configured to generate third representations based on the second representations; an output module configured to select ones of the K blocks based on the third representations; and a summary module configured to generate a summary of the document from text of the selected ones of the K blocks.

Abstract: The present invention relates to a user knowledge tracing method with more improved accuracy, and an operating method for a user knowledge tracing system including a plurality of encoder neural networks and a plurality of decoder neural networks includes: inputting exercise information to a k-th encoder neural network and inputting response information to a k-th decoder neural network; generating query data, which is information on an exercise for which a user is to predict a correct answer probability, by reflecting a weight to the response information and generating attention information to be used as a weight for the query data by reflecting the weight to the exercise information; and training the user knowledge tracing system by using the attention information as the weight for the query data.

Abstract: A method can be used to predict risk using machine learning models having efficient feature learning. A risk prediction model can be applied to time-series data associated with a target entity to generate a risk indicator. The risk prediction model can include a feature learning model for generating features from the time-series data. The risk prediction model can also include a risk classification model for generating the risk indicator. The feature learning model can include filters and can be trained. Parameters of the risk prediction model can be adjusted to minimize a loss function associated with risk indicators. An updated risk prediction model can be generated by removing a filter from an original set of filters based on influencing scores of the original filters. The risk indicator can be transmitted to a computing device for use in controlling access of the target entity to a computing environment.

Abstract: A system is provided. The system includes a first platform including a first platform level agent configured to direct one or more actions of the first platform based on at least one of a selected target or a selected goal. The system also includes a computer system in communication with the first platform level agent. The computer system programmed to a) execute a supervisor level agent configured to select at least one of a target or a goal for one or more platforms including the first platform, b) receive targeting information including one or more targets, c) receive platform information for the one or more platforms, d) select, by the supervisor level agent, a target of the one or more targets based on the target information and the platform information, and e) transmit, to the first platform level agent, the selected target.

Abstract: An information handling system includes a data store configured to store an account associated with a first alert and a second alert. A processor may receive the first alert and the second alert, map the first alert and the second alert to a first node and a second node of a causality graph, traverse the causality graph starting from the sink node to the source node to determine an association between the first alert and the second alert, and generate the account based on the association between the first alert and the second alert.

Abstract: An electronic device includes at least one processor configured to obtain user data associated with a plurality of devices from multiple data sources. The at least one processor is also configured to determine a static weight for each of the plurality of devices based on at least one source of the multiple data sources. The at least one processor is further configured to identify a portion of the plurality of devices that represents the plurality of devices based on the static weight and a dynamic weight. In addition, the at least one processor is configured to determine the dynamic weight for each of the portion of the plurality of devices while the portion of the plurality of devices is identified, where the dynamic weight is based on one or more sources of the multiple data sources.

Abstract: A method, system, and computer program product for managing model updates at multiple data centers hosting a same machine learning model obtain a plurality of first feature profiles input to a first implementation of a first machine learning model and a plurality of first model states determined from processing a model input with the first implementation; determine that a first model policy associated with the first machine learning model is satisfied, based on the plurality of first model states including a set of model states defined by the first model policy and the plurality of first feature profiles including a set of feature profiles defined by the first model policy; and in response to determining that the first model policy is satisfied, provide the plurality of first model states and the plurality of first feature profiles for updating at least one second implementation of the first machine learning model.

Abstract: Methods are provided for deploying machine learning operations within existing storage devices for streamlining various calibration processes. Machine learning operations are specifically designed to generate inference data as a substitute for various measurements taken during calibration. These operations may be verified through additional sample measurements and rolled back when the results of the machine learning operations are outside of a range of approved values. Storage devices designed to utilize machine learning methods within calibration processes can include a non-volatile memory for storing data, executable instructions, and a processor to conduct a variety of steps. The steps can include executing an application stored in the non-volatile memory and receiving a request for measurement data from the application.

Abstract: List-based entity name detection implementations are described that detect entity names in electronic textural documents. In one implementation, unknown entity names are detected. In another implementation, ambiguous entity names are detected and disambiguated. In yet another implementation, generic entity names are detected and associated with an applicable species entity name.

Abstract: One embodiment of the present application sets forth a method for playback of a sense-making operation. The method includes receiving first session data that includes a set of timeline steps. Each timeline step included in the set of timeline steps corresponds to a user action performed on a data set. The method further includes receiving a playback command to display a first sequence of timeline steps included in the set of timeline steps. The method further includes rendering a first graph for display based on at least one timeline step included in the first sequence of timeline steps.

Abstract: A method includes segmenting updates associated with a record into a set of update subsets and generating first and second vectors based on first and second update subsets using a first neural network. The first update subset is associated with a first session and a timestamp, and the second update subset is associated with a second session. The method includes determining a first output using a second neural network based on the first and second vectors and a time difference between the first and second sessions. The method includes selecting a segment of a periodic time interval based on the timestamp, determining a second output using a third neural network based on a ratio based on the segment and the periodic time interval, and generating a characterizing vector using a fourth neural network based on the first and second outputs.