ACSOS 2021
Mon 27 September - Fri 1 October 2021 Washington, DC, United States

A self-learning adaptive system (SLAS) uses machine learning to enable and enhance its adaptability. Such systems are expected to perform well in dynamic situations. For learning high-performance adaptation policy, some assumptions must be made on the environment-system dynamics when information about the real situation is incomplete. However, these assumptions cannot be expected to be always correct, and yet it is difficult to enumerate all possible assumptions. This leads to the problem of incomplete-information learning. We consider this problem as multiple model problem in terms of finding the adaptation policy that can cope with multiple models of environment-system dynamics. This paper proposes a novel approach to engineering the online adaptation of SLAS. It separates three concerns that are related to the adaptation policy and presents the modeling and synthesis process, with the goal of achieving higher model construction efficiency. In addition, it designs a meta-reinforcement learning algorithm for learning the meta policy over the multiple models, so that the meta policy can quickly adapts to the real environment-system dynamics. At last, it reports the case study on a robotic system to evaluate the adaptability of the approach.

Wed 29 Sep

Displayed time zone: Eastern Time (US & Canada) change

11:45 - 12:50
Cross-disciplinary researchMain Track at AUDITORIUM 2
Chair(s): Alessandro Vittorio Papadopoulos Mälardalen University
Timing configurations affect the macro-properties of multi-scale feedback systems
Main Track
Patricia Mellodge University of Hartford, Ada Diaconescu LTCI Lab, Telecom Paris, Institute Politechnqie de Paris, Louisa Jane Di Felice Universidad Autónoma de Barcelona
Causal Inference Techniques for Microservice Performance Diagnosis: Evaluation and Guiding Recommendations
Main Track
Li Wu Elastisys AB/Technische Universität Berlin, Johan Tordsson Elastisys AB, Erik  Elmroth  Elastisys AB/Umea University, Odej  Kao Technische Universität Berlin
A Meta Reinforcement Learning-based Approach for Self-Adaptive System
Main Track
Mingyue Zhang Peking University, China, Jialong Li Waseda University, Japan, Haiyan Zhao Peking University, Kenji Tei Waseda University / National Institute of Informatics, Japan, Shinichi Honiden Waseda University / National Institute of Informatics, Japan, Zhi Jin Peking University