• https://doi.org/10.1145/3319619.3326819Copy DOI Icon

Toward self-learning model-based EAs

  • Jul 13, 2019
  • Erik A Meulman +1 more
Show More
  • Abstract
  • Literature Map
  • References
  • Similar Papers
Abstract

Model-based evolutionary algorithms (MBEAs) are praised for their broad applicability to black-box optimization problems. In practical applications however, they are mostly used to repeatedly optimize different instances of a single problem class, a setting in which specialized algorithms generally perform better. In this paper, we introduce the concept of a new type of MBEA that can automatically specialize its behavior to a given problem class using tabula rasa self-learning. For this, reinforcement learning is a naturally fitting paradigm. A proof-of-principle framework, called SL-ENDA, based on estimation of normal distribution algorithms in combination with reinforcement learning is defined. SL-ENDA uses an RL-agent to decide upon the next population mean while approaching the rest of the algorithm as the environment. A comparison of SL-ENDA to AMaLGaM and CMA-ES on unimodal noiseless functions shows mostly comparable performance and scalability to the broadly used and carefully manually crafted algorithms. This result, in combination with the inherent potential of self-learning model-based evolutionary algorithms with regard to specialization, opens the door to a new research direction with great potential impact on the field of model-based evolutionary algorithms.

Similar Papers
  • PDF
  • Research Article
  • Citations1

An Empirical Investigation on Evolutionary Algorithm Evolving Developmental Timings

  • Nov 06, 2020
  • Electronics
  • Kei Ohnishi +2
  • Conference Article
  • Citations73

A Model-Based Evolutionary Algorithm for Bi-objective Optimization

  • Dec 12, 2005
  • Aimin Zhou +4
  • Conference Article

Model-based evolutionary algorithms

  • Jul 09, 2022
  • Dirk Thierens +1
  • Conference Article

Model-based evolutionary algorithms

  • Jul 06, 2013
  • Dirk Thierens +1
  • Research Article
  • Citations2

A random forest‐regression‐based inverse‐modeling evolutionary algorithm using uniform reference points

  • Oct 01, 2022
  • ETRI Journal
  • Pezhman Gholamnezhad +2
  • Research Article
  • Citations50

Neural Net-Enhanced Competitive Swarm Optimizer for Large-Scale Multiobjective Optimization.

  • Jun 01, 2024
  • IEEE transactions on cybernetics
  • Lingjie Li +6
  • Conference Article
  • Citations3

Reward Shaping in Episodic Reinforcement Learning

  • May 08, 2017
  • Marek Grześ
  • Conference Article
  • Citations6

A Data-Efficient Reinforcement Learning Method Based on Local Koopman Operators

  • Dec 01, 2021
  • Lixing Song +2
  • PDF
  • Research Article
  • Citations4

Feasibility Analysis and Application of Reinforcement Learning Algorithm Based on Dynamic Parameter Adjustment

  • Sep 22, 2020
  • Algorithms
  • Menglin Li +3
  • Research Article
  • Citations18

A 3-phase combined wheel slip and acceleration threshold algorithm for anti-lock braking in heavy commercial road vehicles

  • Mar 22, 2021
  • Vehicle System Dynamics
  • Akhil Challa +5
  • Conference Article
  • Citations40

Hierarchical model-based reinforcement learning

  • Jan 01, 2008
  • Nicholas K Jong +1
  • Supplementary Content

Sample-Efficient I-Projections for Robot Learning

  • Apr 19, 2021
  • TUbilio (Technical University of Darmstadt)
  • Oleg Arenz
  • Conference Article

Inference-based Reinforcement Learning and its Application to Dynamic Resource Allocation

  • Aug 29, 2022
  • Paschalis Tsiaflakis +1
  • Research Article

Reinforcement learning in Connect 4 Game

  • Apr 05, 2021
  • International Journal of Advanced Trends in Computer Science and Engineering
  • Conference Article

Evaluating Reduced Order Models for Training Reinforcement Learning Agents for Building HVAC Control

  • Jul 15, 2024
  • Matthew Kerr +1
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.