Metaheuristics provide the means to approximately solve complex optimisation problems when exact optimisers cannot be utilised. This led to an explosion in the number of novel metaheuristics, most of them metaphor-based, using nature as a source of inspiration. Thus, keeping track of their capabilities and innovative components is an increasingly difficult task. This can be resolved by an exhaustive classification system. Trying to classify metaheuristics is common in research, but no consensus on a classification system and the necessary criteria has been established so far. Furthermore, a proposed classification system can not be deemed complete if inherently different metaheuristics are assigned to the same class by the system. In this paper we provide the basis for a new comprehensive classification system for metaheuristics. We first summarise and discuss previous classification attempts and the utilised criteria. Then we present a multi-level architecture and suitable criteria for the task of classifying metaheuristics. A classification system of this kind can solve three main problems when applied to metaheuristics: organise the huge set of existing metaheuristics, clarify the innovation in novel metaheuristics and identify metaheuristics suitable to solve specific optimisation tasks.

No abstract

No abstract

We present a first evaluation of a new accuracy-based Pittsburghstyle learning classifier system (LCS) for supervised learning of multi-dimensional continuous decision problems: The SupRB-1 (Supervised Rule-Based) learning system. Designed primarily for finding parametrizations for industrial machinery, SupRB-1 learns an approximation of a continuous quality function from examples (consisting of situations, choices and associated qualities-all continous, the first two possibly multi-dimensional) and is then able to make an optimal choice as well as predict the quality of a choice in a given situation. This paper shows and discusses preliminary results of SupRB-1's performance on an additive manufacturing problem.

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.