Contextual CMA-ES

Abdolmaleki, A. and Price, B. and Lau, N. and Reis, P. and Neumann, G. (2017) Contextual CMA-ES. In: International Joint Conference on Artificial Intelligence (IJCAI), 22 - 25 August 2017, Melbourne, Australia.

Documents
Abdolmaleki_IJCA2017.pdf
[img]
[Download]
[img]
Preview
PDF
Abdolmaleki_IJCA2017.pdf - Whole Document

852kB
Item Type:Conference or Workshop contribution (Paper)
Item Status:Live Archive

Abstract

Many stochastic search algorithms are designed to optimize a fixed objective function to learn a task, i.e., if the objective function changes slightly, for example, due to a change in the situation or context of the task, relearning is required to adapt to the new context. For instance, if we want to learn a kicking movement for a soccer robot, we have to relearn the movement for different ball locations. Such relearning is undesired as it is highly inefficient and many applications require a fast adaptation to a new context/situation. Therefore, we investigate contextual stochastic search algorithms
that can learn multiple, similar tasks simultaneously. Current contextual stochastic search methods are based on policy search algorithms and suffer from premature convergence and the need for parameter tuning. In this paper, we extend the well known CMA-ES algorithm to the contextual setting and illustrate its performance on several contextual
tasks. Our new algorithm, called contextual CMAES, leverages from contextual learning while it preserves all the features of standard CMA-ES such as stability, avoidance of premature convergence, step size control and a minimal amount of parameter tuning.

Keywords:contextual stochastic searc, CMA-ES, Multi-Task Learning
Subjects:G Mathematical and Computer Sciences > G760 Machine Learning
Divisions:College of Science > School of Computer Science
Related URLs:
ID Code:28141
Deposited On:04 Aug 2017 15:59

Repository Staff Only: item control page