International Business Machines Corporation

Mayank Agarwal

Yasaman Khazaeni

Sohini Upadhyay

Kartik Talamadupula

Tathagata Chakraborti

Mikhail Yurochkin

Eli M Dow

Michael Muller

Justin David Weisz

John Thomas Richards

Steven I Ross

Stephanie Houde

Kshitij P Fadnis

Djallel Bouneffouf

Fernando Carlos Martinez

Trong Nghia Hoang

Zhongshen Zeng

Soumya Ghosh

Jorge J Barroso Carmona

Borja Godoy

Kristjan Herbert Greenewald

A novel formulation called the Context-Attentive Bandit with Observations (CABO) is described, where only a limited number of features can be accessed by the learner. The present invention is applicable to many problems including problems arising in clinical settings and dialog systems where it is not possible to reveal the whole feature set. The present invention adapts the standard contextual bandit algorithm known as Thompson Sampling with a novel algorithm, we call Context-Attentive Thompson Sampling with Observations (CATSO). Experimental results are included to demonstrate its effectiveness including a regret analysis and an empirical evaluation demonstrating advantages of the disclosed novel approach on several real-life datasets.

Feedback driven decision support in partially observable settings

In federated learning problems, data is scattered across different servers and exchanging or pooling it is often impractical or prohibited. A Bayesian nonparametric framework is presented for federated learning with neural networks. Each data server is assumed to provide local neural network weights, which are modeled through our framework. An inference approach is presented that allows us to synthesize a more expressive global network without additional supervision, data pooling and with as few as a single communication round. The efficacy of the present invention on federated learning problems simulated from two popular image classification datasets is shown.

Bayesian nonparametric learning of neural networks

A computer system adapts processing of expressions by a command-line interface. An expression provided to the command-line interface is analyzed, wherein the command line interface includes pre-defined expression processing. One or more artificial intelligence agents are selected from a plurality of artificial intelligence agents based on the analysis of the expression. The expression is evaluated by the selected one or more artificial intelligence agents to determine processing modifications for the pre-defined expression processing. The expression is processed in accordance with the determined processing modifications and results are provided to the command-line interface. Embodiments of the present invention further include a method and program product for adapting processing of expressions by a shell in substantially the same manner described above.

Flexible artificial intelligence agent infrastructure for adapting processing of a shell

The present disclosure describes systems and methods for a command line interface with artificial intelligence integration. Embodiments of the disclosure provide a command line orchestration component (e.g., including a reinforcement learning model) that provides a generic command line interface environment (e.g., that researchers can interface using a simple sense-act application programming interface (API)). For instance, a command line orchestration component receives commands (e.g., text input) from a user via a command line interface, and the command line orchestration component can identify command line plugins and candidate response from the command line plugins. Further, the command line orchestration component may select a response from the candidate responses based on user preferences, user characteristics, etc., thus providing a generic command line interface environment for various users (e.g., including artificial intelligence developers and researchers).

System and method for the orchestration of plugins on the command line

A feature vector characterizing a system to be analyzed via online partially rewarded machine learning is obtained. Based on the feature vector, a decision is made, via the machine learning, using an online policy. The system is observed for environmental feedback. In at least a first instance, wherein the observing indicates that the environmental feedback is available, the environmental feedback is obtained. In at least a second instance, wherein the observing indicates that the environmental feedback is missing, the environmental feedback is imputed via an online imputation method. the online policy is updated based on results of the obtained environmental feedback and the online imputation method. A decision is output based on the updated online policy.

Online partially rewarded learning

A method of using artificial intelligence to provide source code from an original programming language in a target programming language showing regions of low confidence. The method includes receiving, by a computing device, a code base in an original programming language. The computing device further provides the code base in the original programming language to a target programming language using an artificial intelligence tool. The computing device additionally displays the code base in the target programming language using a visualization tool in a visual interface. The computing device still further displays the regions of uncertainty to a human user in the visual interface. The regions of uncertainty provide low confidence regions of the code base in the target programming language for targeted user intervention. The regions of low confidence correlate with violations to provide displayed actionable insight regions.

Computer assisted programming with targeted visual feedback

Growing community of inventors

Somerville, MA, United States of America

Mayank Agarwal