HyFlex: A Benchmark Framework for Cross-Domain Heuristic Search

Evolutionary Computation in Combinatorial Optimization

Parkes

et al. 2013

Self Cite

Abstract. An apprenticeship-learning-based technique is used as a hyperheuristic to generate heuristics for an online combinatorial problem. It observes and learns from the actions of a known-expert heuristic on small instances, but has the advantage of producing a general heuristic that works well on other larger instances. Specifically, we generate heuristic policies for online bin packing problem by using expert near-optimal policies produced by a hyper-heuristic on small instances, where learning is fast. The "expert" is a policy matrix that defines an index policy, and the apprenticeship learning is based on observation of the action of the expert policy together with a range of features of the bin being considered, and then applying a k-means classification. We show that the generated policy often performs better than the standard best-fit heuristic even when applied to instances much larger than the training set.

Section: Introduction and Related Workmentioning

confidence: 99%

Generalizing Hyper-heuristics via Apprenticeship Learning

Asta

Evolutionary Computation in Combinatorial Optimization

Parkes

et al. 2013

Self Cite

“…HyFlex [14] (Hyper-heuristics Flexible framework) is a cross-domain heuristic search API and HyFlex v1.0 is a software framework written in Java, providing an easy-to-use interface for the development of selection hyper-heuristic search algorithms along with the implementation of several problem domains, each of which encapsulates problem-specific components, such as solution representation and low level heuristics. We will refer to HyFlex v1.0 as HyFlex from this point onward.…”

Section: Introductionmentioning

confidence: 99%

“…HyFlex was initially developed to support the first Crossdomain Heuristic Search Challenge (CHeSC) in 2011 1 . Initially, there were six minimisation problem domains implemented within HyFlex [14]. The HyFlex problem domains have been extended to include three more of them, including 0-1 Knapsack Problem (KP), Quadratic Assignment Problem (QAP) and MaxCut (MAC) [1].…”

Section: Introductionmentioning

confidence: 99%

Performance of Selection Hyper-heuristics on the Extended HyFlex Domains

Almutairi

Communications in Computer and Information Science

Kheiri

et al. 2016

Abstract. Selection hyper-heuristics perform search over the space of heuristics by mixing and controlling a predefined set of low level heuristics for solving computationally hard combinatorial optimisation problems. Being reusable methods, they are expected to be applicable to multiple problem domains, hence performing well in cross-domain search. HyFlex is a general purpose heuristic search API which separates the high level search control from the domain details enabling rapid development and performance comparison of heuristic search methods, particularly hyper-heuristics. In this study, the performance of six previously proposed selection hyper-heuristics are evaluated on three recently introduced extended HyFlex problem domains, namely 0-1 Knapsack, Quadratic Assignment and Max-Cut. The empirical results indicate the strong generalising capability of two adaptive selection hyper-heuristics which perform well across the 'unseen' problems in addition to the six standard HyFlex problem domains.

“…[10,2] are the recent surveys on hyper-heuristics. In this study, a new simple selection hyper-heuristic is introduced and tested across a variety of problem domains using Hyflex (Hyper-heuristics Flexible framework) [11], a software tool for hyper-heuristic development and research.…”

Section: Introductionmentioning

confidence: 99%

A Hyper-heuristic with a Round Robin Neighbourhood Selection

Kheiri

Evolutionary Computation in Combinatorial Optimization

2013

Abstract. An iterative selection hyper-heuristic passes a solution through a heuristic selection process to decide on a heuristic to apply from a fixed set of low level heuristics and then a move acceptance process to accept or reject the newly created solution at each step. In this study, we introduce Robinhood hyper-heuristic whose heuristic selection component allocates equal share from the overall execution time for each low level heuristic, while the move acceptance component enables partial restarts when the search process stagnates. The proposed hyper-heuristic is implemented as an extension to a public software used for benchmarking of hyper-heuristics, namely HyFlex. The empirical results indicate that Robinhood hyper-heuristic is a simple, yet powerful and general multistage algorithm performing better than most of the previously proposed selection hyper-heuristics across six different Hyflex problem domains.