Reinforcement Learning for Online Control of Evolutionary Algorithms

Eiben, A. E.; Horváth, Márk; Kowalczyk, Wojtek; Schut, Martijn C.

doi:10.1007/978-3-540-69868-5_10

Cited by 48 publications

(35 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The only case we are aware of where RL was used for generic parameter control is by Eiben et al [3]. Fitness based metrics were used to define the state while actions were mapped to set any/all parameters.…”

Section: Related Workmentioning

confidence: 99%

Generic parameter control with reinforcement learning

Karafotias¹,

Eiben²,

Hoogendoorn³

2014

Proceedings of the 2014 Annual Conference on Genetic and Evolutionary Computation

View full text Add to dashboard Cite

Parameter control in Evolutionary Computing stands for an approach to parameter setting that changes the parameters of an Evolutionary Algorithm (EA) on-the-fly during the run. In this paper we address the issue of a generic and parameter-independent controller that can be readily plugged into an existing EA and offer performance improvements by varying the EA parameters during the problem solution process. Our approach is based on a careful study of Reinforcement Learning (RL) theory and the use of existing RL techniques. We present experiments using various state-of-the-art EAs solving different difficult problems. Results show that our RL control method has very good potential in improving the quality of the solution found without requiring additional resources or time and with minimal effort from the designer of the application.

show abstract

Section: Related Workmentioning

confidence: 99%

Generic parameter control with reinforcement learning

Karafotias¹,

Eiben²,

Hoogendoorn³

2014

Proceedings of the 2014 Annual Conference on Genetic and Evolutionary Computation

View full text Add to dashboard Cite

show abstract

“…SARSA learning with various reward functions is considered, including combinations of the difference between the current function value and the one evaluated at the last reward computation and the movement in parameter space (the distance traveled in the last phase). On-the-fly parameter tuning, or on-line calibration of parameters for evolutionary algorithms by reinforcement learning (crossover, mutation, selection operators, population size) is suggested in [12]. The EA process is divided into episodes, the state describes the main properties of the current population (like mean fitness -or f values -standard deviation, etc.…”

Section: Reinforcement Learning For Optimizationmentioning

confidence: 99%

An Investigation of Reinforcement Learning for Reactive Search Optimization

Battiti

Campigotto

2011

Autonomous Search

View full text Add to dashboard Cite

Reactive Search Optimization advocates the adoption of learning mechanisms as an integral part of a heuristic optimization scheme. This work studies reinforcement learning methods for the online tuning of parameters in stochastic local search algorithms. In particular, the reactive tuning is obtained by learning a (near-)optimal policy in a Markov decision process where the states summarize relevant information about the recent history of the search. The learning process is performed by the Least Squares Policy Iteration (LSPI) method. The proposed framework is applied for tuning the prohibition value in the Reactive Tabu Search, the noise parameter in the Adaptive Walksat, and the smoothing probability in the Reactive Scaling and Probabilistic Smoothing (RSAPS) algorithm. The novel approach is experimentally compared with the original ad hoc reactive schemes.

show abstract

“…Some reinforcement learning approaches for optimization are also discussed in [8]. Recent work includes [15], on-the-fly parameter tuning for evolutionary algorithms in [55], and the presentation in [14].…”

Section: Reacting On the Objective Functionmentioning

confidence: 99%