Filters








5,795 Hits in 5.0 sec

An overview of LCS research from 2021 to 2022

Michael Heider, David Pätzel, Alexander R. M. Wagner
2022 Proceedings of the Genetic and Evolutionary Computation Conference Companion  
CCS CONCEPTS • Computing methodologies → Rule learning; Genetic algorithms; • General and reference → Surveys and overviews.  ...  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  ...  general rules.  ... 
doi:10.1145/3520304.3533985 fatcat:yf3soktsi5f4rjfxvlhlnt2ljy

Bounding XCS's parameters for unbalanced datasets

Albert Orriols-Puig, Ester Bernadó-Mansilla
2006 Proceedings of the 8th annual conference on Genetic and evolutionary computation - GECCO '06  
This paper analyzes the behavior of the XCS classifier system on imbalanced datasets. We show that XCS with standard parameter settings is quite robust to considerable class imbalances.  ...  We propose a method to estimate the imbalance ratio during XCS's training and adapt XCS's parameters online.  ...  We also thank the support of Enginyeria i Arquitectura La Salle, Ramon Llull University, as well as the support of Ministerio de Ciencia y Tecnología under project TIN2005-08386-C05-04, and Generalitat  ... 
doi:10.1145/1143997.1144250 dblp:conf/gecco/Orriols-PuigB06 fatcat:34nsjlgjsnbzdmqjvxzf32wx7i

A multiple population XCS: Evolving condition-action rules based on feature space partitions

Mani Abedini, Michael Kirley
2010 IEEE Congress on Evolutionary Computation  
In this paper, we investigate the effects of alternative feature space partitioning techniques in a multiple population island-based parallel XCS.  ...  Simulation results show that our multiple population XCS produced better performance and better generalization than the single population XCS model, especially when the problem increased in size.  ...  It might also be interesting to examine a theoretical explanation of the findings. Fig. 1 . 1 XCS model overview.  ... 
doi:10.1109/cec.2010.5586521 dblp:conf/cec/AbediniK10 fatcat:nkwqm5lfozch5osq4yzkkjxidy

Genetics-Based Machine Learning [chapter]

Tim Kovacs
2012 Handbook of Natural Computing  
deletion Introduced in XCS (see [52] ) When rule x subsumes rule y , y is deleted and the numerosity of x incremented In XCS a rule is allowed to subsume another if: It logically subsumes it It is accurate  ...  Helps adapt bias of learner A form of meta-learning Selecting default actions in decision lists GAssist (Pittsburgh) Rulesets are decision lists Initialise rulesets with fully general last rule A default  ... 
doi:10.1007/978-3-540-92910-9_30 fatcat:rm5bx5lwdvfalolrky6lpyt67a

Evolutionary rule-based systems for imbalanced data sets

Albert Orriols-Puig, Ester Bernadó-Mansilla
2008 Soft Computing - A Fusion of Foundations, Methodologies and Applications  
This paper investigates the capabilities of evolutionary on-line rule-based systems, also called learning classifier systems (LCSs), for extracting knowledge from imbalanced data.  ...  The paper adapts and analyzes LCSs for challenging imbalanced data sets and establishes the bases for further studying the combination of re-sampling technique and learner best suited to a specific kind  ...  The authors thank the support of Enginyeria i Arquitectura La Salle, Ramon Llull University, as well as the support of Ministerio de Ciencia y Tecnología under project TIN2005-08386-C05-04, and Generalitat  ... 
doi:10.1007/s00500-008-0319-7 fatcat:iver4ubrkfe2nilt27bcdescxq

A brief history of learning classifier systems: from CS-1 to XCS and its variants

Larry Bull
2015 Evolutionary Intelligence  
The theoretical underpinnings of XCS have also been extended to XCSF (e.g., [Stalph et al., 2012a] ).  ...  This creates a generalization pressure but, importantly, also frees the deletion process of the GA to be used to maintain multiple niches in an emergent way thereby addressing one of the main issues in  ... 
doi:10.1007/s12065-015-0125-y fatcat:tjihtlllsrfhtnohjqehxtkrf4

Combining Gradient-Based With Evolutionary Online Learning: An Introduction to Learning Classifier Systems

Martin V. Butz
2007 7th International Conference on Hybrid Intelligent Systems (HIS 2007)  
Since the introduction of the accuracy-based XCS classifier system by Stewart W.  ...  Holland in the 1970s, are rule-based evolutionary online learning systems that combine gradient-based rule evaluation with evolutionary-based rule structuring techniques.  ...  Generally, an LCS consists of (1) a set of rules, that is, a population of classifiers, (2) a rule evaluation mechanism, and (3) a rule evolution mechanism.  ... 
doi:10.1109/ichis.2007.4344020 fatcat:2ppx7ps6n5d2vbisuuvh2zhfwu

Combining Gradient-Based With Evolutionary Online Learning: An Introduction to Learning Classifier Systems

Martin V. Butz
2007 7th International Conference on Hybrid Intelligent Systems (HIS 2007)  
Since the introduction of the accuracy-based XCS classifier system by Stewart W.  ...  Holland in the 1970s, are rule-based evolutionary online learning systems that combine gradient-based rule evaluation with evolutionary-based rule structuring techniques.  ...  Generally, an LCS consists of (1) a set of rules, that is, a population of classifiers, (2) a rule evaluation mechanism, and (3) a rule evolution mechanism.  ... 
doi:10.1109/his.2007.66 dblp:conf/his/Butz07 fatcat:bhzj5xmimjflzleypxjkyrdmgq

Theoretical Analysis of Accuracy-Based Fitness on Learning Classifier Systems

Rui Sugawara, Masaya Nakata
2022 IEEE Access  
This drawback indicates that XCS may struggle to identify good parent rules at early generations, degrading the efficiency of evolutionary propagation.  ...  The goal of this paper is to complement this fundamental lack in the reliability of XCS by mathematically analyzing its fitness-update scheme.  ...  This fact indicates that XCS suffers in identifying good parent rules in GA, and thus, it hinders an evolutionary propagation at early generations, where the population is fulfilled mostly with baseline  ... 
doi:10.1109/access.2022.3183618 fatcat:ogulz3wtcfd2hl4amlipbufbhy

Learning Classifier Systems: Looking Back and Glimpsing Ahead [chapter]

Jaume Bacardit, Ester Bernadó-Mansilla, Martin V. Butz
2008 Lecture Notes in Computer Science  
In comparison to other machine learning techniques, the advantages of LCSs have become more pronounced: (1) rulecomprehensibility and thus knowledge extraction is straightforward; (2) online learning is  ...  Moreover, we take a glimpse ahead by discussing future challenges and opportunities for successful system applications in various domains.  ...  Fig. 1 . 1 While Michigan-style LCSs evolve one population of rules, in which the rules compete for offspring generation, Pittsburgh-style LCSs evaluate and evolve multiple populations, which compete with  ... 
doi:10.1007/978-3-540-88138-4_1 fatcat:4ywfltzfbzb57a2widrhe5xlam

A Brief History of Learning Classifier Systems: From CS-1 to XCS [article]

Larry Bull
2014 arXiv   pre-print
This paper gives an historical overview of the evolution of such systems up to XCS, and then some of the subsequent developments of XCS to different types of learning.  ...  Modern Learning Classifier Systems can be characterized by their use of rule accuracy as the utility metric for the search algorithm(s) discovering useful rules.  ...  The theoretical underpinnings of XCS have also been extended to XCSF (e.g., [Stalph et al., 2012a] ).  ... 
arXiv:1401.3607v2 fatcat:t4qk5a3ljrhgvmjz7xl5swbtzi

Learning Classifier Systems: A Brief Introduction [chapter]

Larry Bull
2004 Studies in Fuzziness and Soft Computing  
Learning] Classifier systems are a kind of rule-based system with general mechanisms for processing rules in parallel, for adaptive generation of new rules, and for testing the effectiveness of existing  ...  These mechanisms make possible performance and learning without the "brittleness" characteristic of most expert systems in AI. Holland et al., Induction, 1986  ...  Acknowledgements Thanks to everyone involved in this edited collection: Professor Kacprzyk for agreeing to publish the book in his series, Professor Wilson for providing a Foreword, and, of course, the  ... 
doi:10.1007/978-3-540-39925-4_1 fatcat:jagauopuwnawhem46dbg5yu2py

Intrusion detection with evolutionary learning classifier systems

Kamran Shafi, Tim Kovacs, Hussein A. Abbass, Weiping Zhu
2007 Natural Computing  
We detect little sign of overfitting in XCS but somewhat more in UCS. However, both systems tend to reach near-best performance in very few passes over the training data.  ...  Together these two mechanisms enable LCSs to evolve solutions to decision problems in the form of easy to interpret rules called classifiers.  ...  Most of these experiments were run on AC3 supercomputing facilities.  ... 
doi:10.1007/s11047-007-9053-9 fatcat:hwjdqhdph5gq3b3xo3a4apyu2y

Price of Fairness in Kidney Exchange

J. Dickerson, A. Procaccia, T. Sandholm
2014 Transplantation  
We show that the price of fairness in the standard theoretical model is small.  ...  Fielded exchanges typically match under utilitarian or near-utilitarian rules; this approach marginalizes certain classes of patients.  ...  This material was funded by NSF grants IIS-1320620, CCF-1101668, CCF-1215883, and IIS-0964579, by an NDSEG fellowship, and used the Pittsburgh Supercomputing Center in partnership with the XSEDE, which  ... 
doi:10.1097/00007890-201407151-02779 fatcat:njycbov42rardo4s7ksdluxqni

Using the XCS Classifier System for Multi-objective Reinforcement Learning Problems

Matthew Studley, Larry Bull
2007 Artificial Life  
Investigations of how XCS performs in other types of multi-objective learning tasks are also being undertaken, along with comparisons with other approaches [1] .  ...  We are now exploiting these findings to use XCS to control a real mobile robot that must solve such multi-objective problems.  ...  In Section 6 we relate these findings to recent theoretical studies of XCS, and in Section 7 we present our conclusions.  ... 
doi:10.1162/artl.2007.13.1.69 pmid:17204013 fatcat:zhcq7z3csvbypnaksjb2e7shda
« Previous Showing results 1 — 15 out of 5,795 results