Title:
OVERSUBSCRIPTION REINFORCEMENT LEARNER
Document Type and Number:
WIPO Patent Application WO/2024/050824
Kind Code:
A1
Abstract:
A computing system including one or more processing devices that train an oversubscription reinforcement learner at least in part by receiving computing resource usage trajectories. At the oversubscription reinforcement learner, the training further includes generating prototypes based at least in part on the computing resource usage trajectories. The training further includes, based at least in part on the prototypes, generating an oversubscription rate. The training further includes outputting a prototype feedback query and/or an oversubscription rate feedback query. The training further includes receiving a prototype feedback input and/or an oversubscription rate feedback input. Based at least in part on the computing resource usage trajectories, the prototypes, and the prototype feedback input and/or the oversubscription rate feedback input, the training further includes computing an objective function value and training the oversubscription reinforcement learner based at least in part on the objective function value.
More Like This:
Inventors:
WANG LU (US)
DAS MAYUKH (US)
YANG FANGKAI (US)
DONG HANG (US)
QIAO BO (US)
LIU YUDONG (US)
QIN SI (US)
RUEHLE VICTOR JONAS (US)
BANSAL CHETAN (US)
LIN QINGWEI (US)
DAS MAYUKH (US)
YANG FANGKAI (US)
DONG HANG (US)
QIAO BO (US)
LIU YUDONG (US)
QIN SI (US)
RUEHLE VICTOR JONAS (US)
BANSAL CHETAN (US)
LIN QINGWEI (US)
Application Number:
PCT/CN2022/118166
Publication Date:
March 14, 2024
Filing Date:
September 09, 2022
Export Citation:
Assignee:
MICROSOFT TECHNOLOGY LICENSING LLC (US)
WANG LU (CN)
WANG LU (CN)
International Classes:
G06F9/50; G06N3/006; G06N3/092; G06N20/00
Foreign References:
| US20200387394A1 | 2020-12-10 |
Other References:
ALANAZI SULTAN ET AL: "Energy-Aware Resource Management Framework for Overbooked Cloud Data Centers with SLA Assurance", 2018 IEEE GLOBAL COMMUNICATIONS CONFERENCE (GLOBECOM), IEEE, 9 December 2018 (2018-12-09), pages 1 - 6, XP033520042, DOI: 10.1109/GLOCOM.2018.8647884
HO JONATHAN ET AL: "Generative Adversarial Imitation Learning", 10 June 2016 (2016-06-10), pages 1 - 14, XP093022413, Retrieved from the Internet [retrieved on 20230209], DOI: 10.48550/arxiv.1606.03476
HO JONATHAN ET AL: "Generative Adversarial Imitation Learning", 10 June 2016 (2016-06-10), pages 1 - 14, XP093022413, Retrieved from the Internet
Attorney, Agent or Firm:
SHIHUI PARTNERS (CN)
Download PDF:
Previous Patent: METHOD, DEVICE AND COMPUTER PROGRAM PRODUCT FOR WIRELESS COMMUNICATION
Next Patent: COMPOUNDS AS MLKL INHIBITORS
Next Patent: COMPOUNDS AS MLKL INHIBITORS