International Business Machines Corporation

Airbnb, Inc.

Elizabeth Daly

Oznur Alkan

Adi Ionel Botea

Massimiliano Mattetti

Martin Stephenson

Bei Chen

Akihiro Kishimoto

Rahul Nair

Radu Marinescu

Pierpaolo Tommasi

Inge Lise Vejsbjerg

Charles Arthur Jochim

Matthew Scott Davis

Michael Muller

Biplav Srivastava

Yunfeng Zhang

Vera Liao

Anup Kalia

Stefano Braghin

Jin Xiao

Yassine Lassoued

Giovanni Russo

Yufang Hou

Vanessa Lopez Garcia

Spyridon Antonatos

Kartik Talamadupula

Abel Nicolas Valente

Lea Deleris

Stephane Deparis

Pablo Pedemonte

Dennis Wei

Christian Muise

Alan Zwiren

Alice J Chang

Xiaoxi Tian

Maja Vukovic

John Joseph Rofrano

Rachel K E Bellamy

Spyros Kotoulas

Killian Levacher

Karthikeyan Natesan Ramamurthy

Joao H Bettencourt-Silva

Derek Wayne Botti

Mihir Choudhury

Raymond Lloyd

Francesca Bonin

Djallel Bouneffouf

Léa Deleris

Benjamin Dubiel

Lambert Pouguem Wassi

Sanjmeet Abrol

Weiwei Li

Various embodiments are provided for intelligent resolution of conflicting information in a dialog system in a computing environment by a processor. Conflicting information relating to one or more queries may be detected between one or more users using the dialog system. One or more responses may be provided to resolve the conflicting information according to a knowledge domain.

Intelligent resolution of conflicting information in adaptive dialog systems

Embodiments of the invention are directed to a computer-implemented method. A non-limiting example of the computer-implemented method includes accessing, using an explanation generator module of a processor system, information of a recommendation associated with an application, information of the application, and information of a user of the application. The explanation generator module of the processor system is used to determine an explanation format of an explanation of the recommendation based at least in part on the information of the recommendation associated with the application, the information of the application, and the information of the user of the application.

Automatically generating personalized and context-aware explanation formats for a recommendation

A machine learning model can be run with a first input data to obtain a first output from the machine learning model. A first rule representing an interpretable explanation for the machine learning model predicting the first output can be obtained. A modified rule can be received, which changes the first rule. A transformation function can be generated, which transforms a given input data for inputting to the machine learning model to a new input data based on decision boundaries of the machine learning model and the modified rule. The first rule, the first output, the modified rule, and the transformation function can be stored in a first table.

Hybrid user contributed rules and machine learning framework

In a method for interpreting output of a machine learning model, a processor receives a first interpretable rule set. A processor may also receive a second interpretable rule set generated from a dataset and model-predicted labels classifying the dataset. A processor may also generate a difference metric and mapping between the first interpretable rule set and the second interpretable rule set.

Interpretable model changes

Machine learning model change management in an online Software as a Medical Device ('SaMD') is provided. One or more machine learning models implemented in a medical domain may be monitored where the one or more machine learning models are continuously updated. One or more changes to the one or more machine learning models. The one or more machine learning models, having the one or more changes, are certified as being in compliance with performance characteristics and compliance criteria.

Model change management of online software as a medical device

Learning and correcting errors in text input fields to an artificial intelligence (AI) system includes receiving, by the AI system from a target system, an input text. The AI system executes a text processing operation on the input text by applying at least one transformer from a library of transformers to the input text to generate transformed text. At least one proposed correction to the input text is output by the AI system based on an analysis of the transformed text. Feedback data, associated with the at least one proposed correction, is then received from a user of the target system to iteratively learn, by the AI system, which of the transformers need be applied on future input text to accurately generate future proposed corrections on the future input text.

Syntactic and semantic autocorrect learning

A method, system, and computer program product for behavior-based Internet of Things (IoT) device security are provided. The method detects an action from a set of IoT devices. A context is identified for the action and at least one IoT device of the set of IoT devices. The action and the context are validated for the at least one IoT device. The action is identified as an anomaly based on the validating of the action and the context. A potential state change is identified for the at least one IoT device based on the anomaly. The method determines a responsive action based on the potential state change and the anomaly.

Behavior driven security for IoT devices

Embodiments are disclosed for a method. The method includes receiving feedback decision rules for multiple predictions by a trained machine learning model. generating a feedback rule set based on the feedback decision rules. The method further includes generating an updated training dataset based on an original training dataset and an updated feedback rule set. The updated feedback rule set resolves one or more conflicts of the feedback rule set, and the updated training dataset is configured to train the machine learning model to move a decision boundary. Generating the updated training dataset includes generating multiple updated training instances by applying one of the feedback decision rules to a training instance of the original training dataset.

Moving decision boundaries in machine learning models

An online machine learning model such as an autonomous agent predicts an action. A processor associated with, or running, the online machine learning model observes an environment for an interval of time for a real reward associated with the action. Responsive to determining that the real reward is not received within the interval of time, the processor determines based on a criterion whether to allocate an immediate reward received within the interval of time to the online machine learning model, where the immediate reward is an approximation of the real reward. Responsive to determining that the immediate reward is to be allocated, the processor allocates the immediate reward to the online machine learning model. The online machine learning model further learns or retrains itself based on the immediate reward.

Online machine learning with immediate rewards when real rewards are delayed

Techniques regarding discovering configuration information for one or more computer applications are provided. For example, one or more embodiments described herein can comprise a system, which can comprise a memory that can store computer executable components. The system can also comprise a processor, operably coupled to the memory, and that can execute the computer executable components stored in the memory. The computer executable components can comprise a configuration component that can discover configuration information associated with a containerized computer application. The configuration information can be characterized by a set of environment attributes extracted by querying a source code of the containerized computer application.

Configuration discovery of computer applications

Embodiments for entity transaction interaction analysis and summarization by a processor. Transaction elements relating to one or more entity transaction interactions may be identifies and extracted from one or more communications. The transaction elements may be combined with one or more transaction opportunities and transaction historical data to provide a transaction summary.

Transaction interaction analysis and summarization

Systems, computer-implemented methods and/or computer program products that facilitate answering questions that involve mathematical computations are provided. In one embodiment, a computer-implemented method comprises: transforming, by a system operatively coupled to a processor, a natural language query into a first logical representation and extrinsic knowledge into a second logical representation relevant to the natural language query; merging, by the system, the first logical representation and the second logical representation into a third logical representation; and generating, by the system, answers for the natural language query based on processing of the third logical representation.

Facilitating answering questions involving reasoning over quantitative information

Embodiments for recommending actions to improve machine learning predictions by a processor. One or more recommended actions may be linked to one or more features that influence a predicted outcome of a prediction model of a machine learning operation. One or more features having one or more negative factors that negatively impact the predicted outcome of the prediction model may be determined and selected. One or more of the linked recommended actions may be applied to one or more of the features to mitigate a negative impact upon the predicted outcome of the prediction model.

Linking actions to machine learning prediction explanations

Embodiments of the invention are directed to techniques that include predicting, by a computer system, a number of predicted opportunities and signatures of the predicted opportunities expected in a time window. Based on the signatures of the predicted opportunities, the computer system generates a listing of entities ranked according to signatures of the predicted opportunities. The computer system selects the entities to be assigned to the predicted opportunities based, at least in part, on computing capacity related to sales while accounting for any current opportunities having been assigned to the entities.

Predictive capacity optimizer

A method incrementally solves a current journey planning request from a user. The solving step performs a current search for at least one journey plan that satisfies the request by accessing a database storing journey planning information derived from results to a plurality of previous requests. The solving step stores, in the database, information discovered during the current search for responding to a subsequent request. The information discovered during the current search for responding to the request includes a reusable portion of a search graph, pairs of a state and a lower bound on a best arrival time and pairs of a state and an exact value for the arrival time. The lower bound is employed to increase an accuracy of a pre-computer heuristic function which guides the search based on state dominance in search spaces in which heuristic values are back propagated and stored in the database.

Incremental search based multi-modal journey planning

Growing community of inventors

Dublin, Ireland

Elizabeth Daly