Q-Learning-Based Adaptive Bacterial Foraging Optimization

Ben Niu, Bowen Xue

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review


As a common biological heuristic algorithm, Bacterial Foraging Optimization (BFO) is often used to solve optimization problems. Aiming at increasing the solution accuracy and convergence performance while enhancing the capability of individual’s self-learning and exploration, a Q-Learning-Based Adaptive Bacterial Foraging Optimization (QABFO) is proposed in this paper. The basic chemotaxis, reproduction and elimination/dispersal operations in standard BFO are redesigned in a Q-learning mechanism, and the Q-table will be updated on the basis of the changed fitness values in each iteration. In chemotaxis operation, modified search direction and secondary cruising mechanism are introduced into tumbling and swimming behaviors, with the purpose of improving the search efficiency and balancing local and global search. Additionally, a generation skipping adaptive reproduction is designed to control the accuracy and convergence of QABFO. Experimental results demonstrate that compared with BFO, PSO and GA, the proposed algorithm performs better in terms of accuracy and stability on most of the test functions and can effectively improve the premature convergence problem due to the original reproduction operation in BFO.

Original languageEnglish
Title of host publicationMachine Learning for Cyber Security
Subtitle of host publicationThird International Conference, ML4CS 2020, Guangzhou, China, October 8–10, 2020, Proceedings, Part II
EditorsXiaofeng Chen, Hongyang Yan, Qiben Yan, Xiangliang Zhang
Place of PublicationCham
PublisherSpringer Cham
Number of pages11
ISBN (Electronic)9783030624606
ISBN (Print)9783030624590
Publication statusPublished - 11 Nov 2020
Event3rd International Conference on Machine Learning for Cyber Security, ML4CS 2020 - Guangzhou, China
Duration: 8 Oct 202010 Oct 2020

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference3rd International Conference on Machine Learning for Cyber Security, ML4CS 2020


  • bacterial foraging optimization
  • generation skipping adaptive reproduction
  • Q-learning


Dive into the research topics of 'Q-Learning-Based Adaptive Bacterial Foraging Optimization'. Together they form a unique fingerprint.

Cite this