International Business Machines Corporation

Kyndryl, Inc.

Airbnb, Inc.

Seema Nagar

Kuntal Dey

Roman Vaculin

Karan Ahuja

Pramod Vadayadiyil Raveendran

Enara C Vijil

Sougata Mukherjea

Manish Anand Bhide

Sudhanshu Shekhar Singh

Swati Rallapalli

Vijay Ekambaram

Pramod Verma

Pooja Aggarwal

Rachel Leekin

Saiprasad Kolluri Venkata Sesha

Sameep Mehta

Vinayak Sastri

Srikanth Govindaraj Tamilselvam

Amit Anil Nanavati

James David Cleaver

Abhijit Mishra

Michael James McGuire

Prateek Goyal

Utkarsh Dwivedi

Amitkumar Manoharrao Paradkar

Sumit Mittal

Dipanjan Chakraborty

Koustuv Dasgupta

Willie L Scott, Ii

Nishtha Madaan

Rohan R Arora

Thuy Luong

Charu Pandhi

Meenal Kapoor

Heena Bansal

Sarbajit Kumar Rakshit

Larisa Shwartz

Nilanjan Banerjee

Natwar Modani

Josephine E Justin

Jayakarthik Jayabalan

Abhishek Kaul

Rizwan Sheikh Abdulla

Konda Reddy Dumpa

Saritha Arunkumar

Danish Contractor

Pradipta De

Harshit Kumar

Madhavi Katari

Jitendra Singh

Anupam Joshi

Mukul Tewari

Sreekanth L Kakaraparthy

Ruchika Banerjee

Angshu Rai

Ruta Mehta

Sachin Gupta

Dipanwita Guhathakurta

Lisa Seacat DeLuca

Jonathan D Dunne

Jeremy A Greenberger

Venkata Vara Prasad Karri

Ruchi Mahindru

Mudhakar Srivatsa

Tamer E Abuelsaad

Jinho Hwang

Mukesh Kumar Mohania

Vijay A Kumar

Hemant Kumar Sivaswamy

Diptikalyan Saha

Ramasuri Narayanam

Anbang Xu

L Venkata Subramaniam

Steven A Keller

Arvind Agarwal

Rema Ananthanarayanan

Raghavendra Singh

Laxmikantha Sai Nanduru

Girish Bhimrao Chafle

Amir Ronen

Joshua M Rosenkranz

Seep Goel

Ian Manning

Xiaotong Liu

Qing Wang

Kim Poh Wong

Ilgen Banu Yuceer

Anil Kumar Narigapalli

Matthew Richard James Thornhill

Atul Kumar

Bing Zhou

Hemank Lamba

Kartik Srinivasan

Vijender Koorella

Joydeep Mondal

Sahil Bansal

Subramanya Kabbinahithlu Narayana Prasad

Noah Zheutlin

Stephen James Hussey

Pooja Aggarwall

A method includes receiving, by a computing device, concepts of a domain; determining, by the computing device, objects relevant to the concepts; generating, by the computing device, a new image by stitching the relevant objects together; determining, by the computing device, whether the new image is accurate or inaccurate; and in response to determining the new image is inaccurate, propagating, by the computing device, the inaccurate new image back to a convolutional neural network (CNN).

Object stitching image generation

A system, method, and computer program product for implementing artificial intelligence module communication is provided. The method includes generating models associated with communications between real-time artificial intelligence modules. Classifiers associated with the models are generated and a master real-time artificial intelligence module associated with the modules and classifiers is detected. Real-time artificial intelligence modules are detected and communications between the master real-time artificial intelligence module and the real-time artificial intelligence modules are initiated. Updated classifiers, updated models, and updated self learning software are generated. The updated self learning software is executed and a resulting query associated with a Web search is executed.

Artificial intelligence module communication

Provided is a computer-implemented method, system, and computer program product for dynamically configuring a virtual private network (VPN) protocol of a VPN tunnel. A processor may analyze data activity associated with a first device, where the first device is connected to a second device using a VPN tunnel. The processor may compare the analyzed data activity to a set of policies for determining a VPN protocol to apply to the VPN tunnel. The processor may identify, based on the comparing, a first VPN protocol from a plurality of VPN protocols. The processor may apply the first VPN protocol to the VPN tunnel.

Dynamic virtual private network protocol configuration

Provided is a method, system, and computer program product for storing images across multiple distributed computing systems according to image sensitivity. The method comprises identifying an image and analyzing the image to identify sensitive information in the image. The method further comprises splitting the image into a sensitive portion and a non-sensitive portion. The method further comprises storing the sensitive portion of the image in a first distributed computing system and storing the non-sensitive portion of the image in a second distributed computing system.

Image storage across distributed computing systems

Techniques are provided for dynamic alert suppression policy management. In one embodiment, the techniques involve receiving an event stream, wherein the event stream includes metric values comprising at least one of: log anomaly data and metric anomaly data, determining an anomalous event based on the event stream, determining a persistent region of the anomalous event, determining a quantum representation of the persistent region, determining X-Y values of the persistent region based on the quantum representation, and generating a policy based on a set of the X-Y values.

Dynamic alert suppression policy management

Localizing a faulty microservice in a microservice architecture is achieved by developing healthy execution sequence data for comparison to execution sequences during system failures. Oftentimes the faulty microservice does not emit a failure signal. Frequent sub-sequences arising from log template time series data during healthy execution facilitates localization of faulty services when there is no failure signal from the faulty service.

Fault localization in a distributed computing system

An approach is provided for determining a carbon footprint-based consumption of cloud resources. Cloud resource requirements and a carbon footprint cap of a workload of a cloud consumer are received. The requirements and cap are based on a sustainability target, published by the cloud consumer, and subscribed by cloud service providers. A list of cloud resources that satisfy the requirements are sent. Carbon emission values of the cloud resources at different load levels of the workload are sent. Based on a service level agreement requirement, a criticality level, and a peak load duration of the workload, and previous success rates of satisfying cloud resource requirements by cloud service providers, an optimized configuration of cloud resource(s) and cloud service provider(s) is selected. The cloud resource(s) are selected from the list and have a carbon footprint that does not exceed the cap at a given load level.

Carbon footprint-based control of cloud resource consumption

A computer-implemented method and a computer system for identifying persistent anomalies for failure prediction. The computer system receives a time series data stream. The computer system received a predetermined number N and a predetermined number M which is a fraction of N. The computer system segments the time series data stream into N consecutive sliding windows. The computer system performs supervised persistent anomaly detection to determine whether anomalies across the N consecutive sliding windows are persistent, by using a binary classification model. The computer system performs unsupervised persistent anomaly detection to determine whether the anomalies across the N consecutive sliding windows are persistent. The computer system combines results of the supervised persistent anomaly detection and results of the unsupervised persistent anomaly detection to determine persistent anomalies.

Identifying persistent anomalies for failure prediction

Systems, methods and/or computer program products maximizing answer coherence of virtual agents responding to multi-part user queries, using relationships among extracted chunks corresponding to support document(s). Relationships are extracted by mining entities from text and images of support document(s) and mapping entities to an otology. Induced subgraphs for chunks, based on extracted entities are created and compared. Depending on amounts of overlap and/or relationships between subgraphs, relationships among chunks are identified. Chunks relationships can be complementary, contained within one another, instances of one another, or not related at all. For a given multi-part query, a query answer graph is created comprising answer nodes corresponding to potential answer candidates associated with each chunk. Using the edges between nodes and relationships identified between chunks, coherence scores are calculated along answer pathways, whereby answer pathways with the highest coherence score are selected to respond to the query.

Multi-chunk relationship extraction and maximization of query answer coherence

An unmanned vehicle includes a body, a propulsion system connected to the body, a computing system connected to the body, and a display connected to an exterior of the body and electrically connected to the computing system. In addition, the computing system is configured to dynamically change the appearance of the display during operation of the unmanned vehicle.

Differentiating unmanned vehicles by changing exterior appearance

An embodiment includes detecting a set of anomalies recorded during a first predefined window of time in log entries for a computer environment. The embodiment also includes generating cluster data representative of a cluster of anomalies from among the set of anomalies, where the cluster is formed using a lattice clustering algorithm that spatially distinguishes the cluster of anomalies from other anomalies in the set of anomalies. The embodiment also includes composing an explanation using log templates generated from log entries associated with the cluster of anomalies.

Generating incident explanations using spatio-temporal log clustering

Detecting system log anomalies by receiving multivariate time-series system log data an multivariate metric data, inferring system metrics from the system log data and metric data, receiving a metric causal graph including causal relationships between system metrics, determining a univariate variation score for the system metrics, determining a causal variation score for the multivariate time series system metric data, according to the causal graph, determining an overall activity score according to the univariate variation score, and causal variation score, and altering a review window duration according to the activity score.

Dynamic window-size selection for anomaly detection

A computer-implemented method and a computer system for detecting and preventing distributed data exfiltration attacks. The computer system calculates historical usage statistics for a service, instances of the service, and clients requesting the instances, generates a baseline of normal usage activities for the clients and the instances based on the historical usage statistics, monitors current activities of the clients to build signatures of queries by the clients and signatures of the instances, and correlates the signatures to determine whether a data exfiltration attack is in progress. In response to determining that the data exfiltration attack is in progress, the computer system increases one or more risk scores corresponding to the data exfiltration attack. In response to determining that the one or more risk scores and an overall risk score of the service exceed a predetermined threshold, the computer system generates an alert of the data exfiltration attack.

Detecting and preventing distributed data exfiltration attacks

Provided is a method, computer program product, and system for automatically assigning robotic devices to users based on need using predictive analytics. A processor may monitor activities performed by one or more users. The processor may determine, based on the monitoring, a set of activities that require assistance from a robotic device when being performed by the one or more users. The processor may match the set of activities to a set of capabilities related to a plurality of robotic devices. The processor may identify, based on the matching, a first robotic device that is capable of assisting the one or more users in performing a first activity of the set of activities. The processor may deploy the first robotic device to assist the one or more users in performing the first activity.

Assignment of robotic devices using predictive analytics

Methods, computer program products, and systems are presented. The methods include, for instance: training of a machine learning model for predicting a rating in certain content rating systems based on training data. The machine learning model includes a plurality of maturity classifiers corresponding to individual features of the previously rated contents. An input content is obtained and features of the input content are extracted by use of respective content analysis tools and subsequently classified by the maturity classifiers of the machine learning model.

Dynamic content rating assistant

Growing community of inventors

Bengaluru, India

Seema Nagar