A New Algorithm for Adaptive Online Selection of Auxiliary Objectives

Research output: Chapter in Book/Report/Conference proceedingConference Proceeding (Non-Journal item)

Abstract

Consider optimization problems, where a target objective should be optimized. Some auxiliary objectives can be used to obtain the optimum of the target objective in less number of objective evaluations. We call such auxiliary objective a supporting one. Usually there is no prior knowledge about properties of auxiliary objectives, some objectives can be obstructive as well. What is more, an auxiliary objective can be both supporting and obstructive at different stages of the target objective optimization. Thus, an adaptive online method of objective selection is needed. Earlier, we proposed a method for doing that, which is based on reinforcement learning. In this paper, a new algorithm for adaptive online selection of optimization objectives is proposed. The algorithm meets the interface of a reinforcement learning agent, so it can be fit into the previously proposed framework. The new algorithm is applied for solving some benchmark problems with single-objective evolutionary algorithms. Specifically, Leading Ones with OneMax auxiliary objective is considered, as well as the MH-IFF problem. Experimental results are presented. The proposed algorithm outperforms Q-learning and random objective selection on the considered problems.
Original languageEnglish
Title of host publicationICMLA '14
Subtitle of host publicationProceedings of the 2014 13th International Conference on Machine Learning and Applications
EditorsC. Ferri, G. Qu, X. Chen, M. A. Wani, P. Angelov, J.-H Lai
PublisherIEEE Press
Pages584-587
Number of pages4
ISBN (Electronic)978-1-4799-7415-3
DOIs
Publication statusPublished - 03 Dec 2014
Externally publishedYes
Event2014 13th International Conference on Machine Learning and Applications (ICMLA) - Detroit, United States of America
Duration: 03 Dec 201406 Dec 2014

Conference

Conference2014 13th International Conference on Machine Learning and Applications (ICMLA)
Country/TerritoryUnited States of America
CityDetroit
Period03 Dec 201406 Dec 2014

Keywords

  • reinforcement learning
  • multi-objectivization
  • evolutionary algorithms
  • parameter control

Fingerprint

Dive into the research topics of 'A New Algorithm for Adaptive Online Selection of Auxiliary Objectives'. Together they form a unique fingerprint.

Cite this