XCS Revisited: A Novel Discovery Component for the eXtended Classifier System

Fredivianus, Nugroho; Prothmann, Holger; Schmeck, Hartmut

doi:10.1007/978-3-642-17298-4_30

Cited by 11 publications

(3 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In XCS, a subsumption mechanism combines similar rules, and a randomised deletion mechanism removes classifiers of a low fitness from the population. In (Fredivianus et al, 2010), the discovery component is altered by introducing a modified rule combining technique. The goal is to create maximally general classifiers that match as many inputs as possible while still being exact in their predictions.…”

Section: State Of the Art In Active Reinforcement Learningmentioning

confidence: 99%

Active Reinforcement Learning -- A Roadmap Towards Curious Classifier Systems for Self-Adaptation

Reichhuber¹,

Tomforde²

2022

Preprint

View full text Add to dashboard Cite

Intelligent systems have the ability to improve their behaviour over time taking observations, experiences or explicit feedback into account. Traditional approaches separate the learning problem and make isolated use of techniques from different field of machine learning such as reinforcement learning, active learning, anomaly detection or transfer learning, for instance. In this context, the fundamental reinforcement learning approaches come with several drawbacks that hinder their application to real-world systems: trial-and-error, purely reactive behaviour or isolated problem handling. The idea of this article is to present a concept for alleviating these drawbacks by setting up a research agenda towards what we call "active reinforcement learning" in intelligent systems.

show abstract

Section: State Of the Art In Active Reinforcement Learningmentioning

confidence: 99%

Active Reinforcement Learning -- A Roadmap Towards Curious Classifier Systems for Self-Adaptation

Reichhuber¹,

Tomforde²

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…RXCS-RC implements all concepts as the binary-mode XCS-RC (see Figure 2 and [1]). They mainly modify the discovery component of XCS, with an addition of a new deletion mechanism.…”

Section: The Rxcs-rcmentioning

confidence: 99%

“…A variant of XCS called XCS-RC, which uses the Rule Combining (RC) technique, has successfully solved some tasks using binary inputs for both single-step and multistep mode [1]. Here, the investigated system (denoted as Real-value XCS-RC or RXCS-RC afterwards) is learning to generalize classifiers by creating new rules based on conclusions taken from the already existing rules.…”

Section: Introductionmentioning

confidence: 99%

Stay real!

Fredivianus

Kara

Schmeck

2012

Proceedings of the 14th Annual Conference Companion on Genetic and Evolutionary Computation

Self Cite

View full text Add to dashboard Cite

This paper provides an extension of the rule combining (RC) technique in the Accuracy-based Learning Classifier System (XCS) to handle continuous-valued input. Previously implemented to cope with binaries, the suitability of the newly introduced algorithm is investigated in further tasks. Several experiments are run and the results are compared to previous work using the real-valued multiplexer problem. The comparison shows that by implementing the RC technique, real value XCS is capable of producing a compact population of rules through proper generalizations. Moreover, the learning rate of Real-value XCS-RC (RXCS-RC) is comparable or even superior, in some cases.

show abstract