International Business Machines Corporation

Mahadev Khapali

Shashank Vijaykumar Vagarali

Santosh Hegde

Mohan Narayanswamy

Yugandhra Rayanki

Dattatreya Govindappanavar

Samar T Desai

Hemant Singh

Charan Acharya Chandrashekar

Prabhu S Padashetty

A method for receiving an end-user model access data set, deriving a plurality of patterns of actions typically performed by the end-user based on analysis of the end-user model access data set, and deriving a first model deployment protocol to automatically deploy selected ML models of the plurality of ML models for the end-user when the end-user works with ML models based on the plurality of patterns of actions.

Managing user machine learning (ML) models

A computer-implemented method for creating an auto-scaled predictive analytics model includes determining, via a processor, whether a queue size of a service master queue is greater than zero. Responsive to determining that the queue size is greater than zero, the processor fetches a count of requests in a plurality of requests in the service master queue and a type for each of the requests. The processor derives a value for time required for each of the requests and retrieves a number of available processing nodes based on the time required for each of the requests. The processor then auto-scales a processing node number responsive to determining that a total execution time for all of the requests in the plurality of requests exceeds a predetermined time value and outputs an auto-scaled predictive analytics model based on the processing node number and queue size.

Auto scaling a distributed predictive analytics system with machine learning

In an approach to deriving highly accurate models, one or more computer processors train a set of machine learning models utilizing a training set and a deep learning algorithm; generate one or more feedback data sets for each model in the set of trained models; rank each model in the set of trained models based on the generated feedback data sets; dynamically adjust one or more thresholds, that initiate a retraining or deployment of one or more ranked models, based, at least in part, on one or more production environment requirements; responsive to exceeding one or more adjusted thresholds, automatically deploy one or more ranked models to one or more deployment environments based, at least in part, on the ranking of the one or more trained models; responsive to not exceeding one or more adjusted thresholds, retrain each model in the set of trained models.

Intelligent retraining of deep learning models utilizing hyperparameter sets

A method for updating a plurality of database statistics during a query execution is provided. The method may include receiving a compiled query. The method may also include determining if the received compiled query requires at least one table scan operation on a full table to resolve the received compiled query. The method may further include determining if a plurality of table statistics associated with the full table has not been updated within a pre-defined time-period, whereby the determining is based on the determination that the received compiled query requires at least one table scan operation on the full table to resolve the received compiled query. The method may additionally include collecting a plurality of statistics associated with the full table based on the determination that the plurality of table statistics associated with the full table has not been updated within the pre-defined time-period.

Updating database statistics during query execution

A computer-implemented method includes identifying a mobile phone. The method includes identifying one or more input devices. The one or more input devices are associated with the mobile phone. The method includes collecting behavior information from the input devices. The method includes applying machine learning to the behavior information to yield a schedule.

Scheduling telephone calls

An approach for improving application performance after database recovery is provided, the approach involving tracking one or more applications connecting to a database, tracking metadata in memory on a server computer, wherein the metadata is accessed by the one or more applications, recovering the metadata during a database recovery occurring after a database crash and repopulating the metadata in memory on the server computer during the database recovery, wherein the repopulating occurs prior to the metadata being requested by the one or more applications.

Growing community of inventors

Bengaluru, India

Mahadev Khapali