An overview of LCS research from 2021 to 2022

Abstract

The International Workshop on Learning Classifier Systems (IWLCS) is an annual workshop at the GECCO conference where new concepts and results regarding learning classifier systems (LCSs) are presented and discussed. One recurring part of the workshop agenda is a presentation that reviews and summarizes the advances made in the field over the last year; this is intended to provide an easy entry point to the most recent progress and achievements. The 2020 and 2021 presentations were accompanied by survey workshop papers, a practice which we hereby continue. We give an overview of all the LCS-related publications from 11 March 2021 to 10 March 2022. The 42 publications we review are grouped into six overall topics: Formal theoretic advances, new LCS architectures, LCS-based reinforcement learning, algorithmic improvements to existing LCSs, combinations of LCS and Deep Learning models and, finally, a variety of applications of LCSs. CCS CONCEPTS• Computing methodologies → Rule learning; Genetic algorithms; • General and reference → Surveys and overviews.

An Online Rule Set Compaction Strategy for the sUpervised Classifier System UCS

“…1 ID: is the abbreviation of each data set name. 2 NInst is the number of instances in each data set. 3 Att is the number of attributes.…”

“…LCSs yield to an entire population of rules which have the ability to learn iteratively and collaborate to solve a given problem. One of the major factors of the LCSs success is their rule-based representation [2][3][4] . They combine generalization mechanisms and genetic algorithm with the aim of obtaining a set of maximally general and accurate rules which together define the decision boundaries.…”

“…The objective is to discover homogeneous clusters within unlabeled training data.The given instance is assigned to the cluster with the most similar center 3 . Statistics have revealed that LCSs achieved a high performance at solving these different types of machine learning problems 2,8 . Nevertheless, the high competence of LCSs has been impaired to some extent by the large sets of overlapped rules, where many of them are redundant or poor 9,10 .…”