Contrasting meta-learning and hyper-heuristic research: the role of evolutionary algorithms

Abstract

The fields of machine meta-learning and hyper-heuristic optimisation have developed mostly independently of each other, although evolutionary algorithms (particularly genetic programming) have recently played an important role in the development of both fields. Recent work in both fields shares a common goal, that of automating as much of the algorithm design process as possible. In this paper we first provide a historical perspective on automated algorithm design, and then we discuss similarities and differences between meta-learning in the field of supervised machine learning (classification) and hyper-heuristics in the field of optimisation. This discussion focuses on the dimensions of the problem space, the algorithm space and the performance measure, as well as clarifying important issues related to different levels of automation and generality in both fields. We also discuss important research directions, challenges and foundational issues in meta-learning and hyper-heuristic research. It is important to emphasize that this paper is not a survey, as several surveys on the areas of meta-learning and hyper-heuristics (separately) have been previously published. The main contribution of the paper is to contrast meta-learning and hyper-heuristics methods and concepts, in order to 123Genet Program Evolvable Mach DOI 10.1007/s10710-013-9186-9 promote awareness and cross-fertilisation of ideas across the (by and large, nonoverlapping) different communities of meta-learning and hyper-heuristic researchers. We hope that this cross-fertilisation of ideas can inspire interesting new research in both fields and in the new emerging research area which consists of integrating those fields.

Evolutionary model building under streaming data for classification tasks: opportunities and challenges

“…This then makes the design of systems for streaming data tasks especially challenging. One potential path for addressing this, albeit not yet considered in the wider literature, is through evolving hyper-heuristics (or the evolution of the evolutionary algorithms), e.g., [148]. Current research has not considered this avenue, so we revisit the opportunity in the context of future research (Sect.…”

Distribution of waiting time for dynamic pickup and delivery problems

“…As Pappa et al (2013) point out, it can be generally observed, that research efforts in different fields including operations research, optimization, and machine learning ultimately evolved from algorithm / parameter selection and control to the automated generation of algorithms. The combination of simulation-based optimization and evolutionary computation allows performing a search in the space of possible policies by trial and error interactions with a simulation model (Whiteson 2012).…”

Investigating fitness functions for a hyper-heuristic evolutionary algorithm in the context of balanced and imbalanced data classification

“…Nevertheless, hyper-heuristics usually employ typical metaheuristics (e.g., evolutionary algorithms) as the search methodology to look for suitable heuristics to a computational problem [31].…”