International Business Machines Corporation

Priya Ashok Nagpurkar

Parijat Dube

Gauri Joshi

Jose Eduardo Moreira

Zehra Noman Sura

Diana Jeanne Arroyo

Stefania Costache

Hung Qui Le

Maged M Michael

Naresh Nayar

Harold Wade Cain, Iii

Randal Craig Swanberg

Bradly George Frey

Cathy May

David Michael Daly

Shripad Nadgowda

Benjamin Herrenschmidt

Kaoutar El Maghraoui

Jessica Hui-Chun Tseng

David Joel Edelsohn

Sanghamitra Dutta

Techniques for estimating runtimes of one or more machine learning tasks are provided. For example, one or more embodiments described herein can regard a system that can comprise a memory that stores 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 an extraction component that can extract a parameter from a machine learning task. The parameter can define a performance characteristic of the machine learning task. Also, the computer executable components can comprise a model component that can generate a model based on the parameter. Further, the computer executable components can comprise an estimation component that can generate an estimated runtime of the machine learning task based on the model. The estimated runtime can define a period of time beginning at an initiation of the machine learning task and ending at a completion of the machine learning task.

Runtime estimation for machine learning tasks

A method for performing machine learning includes assigning processing jobs to a plurality of model learners, using a central parameter server. The processing jobs includes solving gradients based on a current set of parameters. As the results from the processing job are returned, the set of parameters is iterated. A degree of staleness of the solving of the second gradient is determined based on a difference between the set of parameters when the jobs are assigned and the set of parameters when the jobs are returned. The learning rates used to iterate the parameters based on the solved gradients are proportional to the determined degrees of staleness.

Adaptive learning rate schedule in distributed stochastic gradient descent

Embodiments relate to a system, program product, and method for evaluating and controlling configuration of a build manifest. An application build manifest is discovered and is subjected to parsing process in which one or more components that comprise the application are identified. The build manifest is monitored for changes to the identified components, and a change notification is generated in response to a change in an identified component. Each generated change notification is assigned a classification. The change notifications are applied selectively to update the manifest, wherein the selective update is based on the classification of the change notification.

Federated framework for container management

A method includes embedding clothing with at least one sensor and at least one control unit; a power unit powering on the at least one sensor and the at least one control unit; the at least one sensor monitoring a sensed condition; the at least one control unit conducting a heat prediction based on the sensed condition; and the at least one control unit controlling threads within the clothing based on the heat prediction to actively adjust properties of the clothing.

Actively controlled performance clothing

Mechanisms are provided, in a data processing system having a processor and a transactional memory, for executing a transaction in the data processing system. These mechanisms execute a transaction comprising one or more instructions that modify at least a portion of the transactional memory. The transaction is suspended in response to a transaction suspend instruction being executed by the processor. A suspended block of code is executed in a non-transactional manner while the transaction is suspended. A determination is made as to whether an interrupt occurs while the transaction is suspended. In response to an interrupt occurring while the transaction is suspended, a transaction abort operation is delayed until after the transaction suspension is discontinued.

Transactional memory system supporting unbroken suspended execution

Growing community of inventors

Ossining, NY, United States of America

Priya Ashok Nagpurkar