pith. sign in

arxiv: 1606.06041 · v1 · pith:AWCY6ONVnew · submitted 2016-06-20 · 💻 cs.AI · cs.NE

Bandit-Based Random Mutation Hill-Climbing

classification 💻 cs.AI cs.NE
keywords algorithmhill-climbingmutationrandomneighbourbandit-baseddiscretenoise-free
0
0 comments X
read the original abstract

The Random Mutation Hill-Climbing algorithm is a direct search technique mostly used in discrete domains. It repeats the process of randomly selecting a neighbour of a best-so-far solution and accepts the neighbour if it is better than or equal to it. In this work, we propose to use a novel method to select the neighbour solution using a set of independent multi- armed bandit-style selection units which results in a bandit-based Random Mutation Hill-Climbing algorithm. The new algorithm significantly outperforms Random Mutation Hill-Climbing in both OneMax (in noise-free and noisy cases) and Royal Road problems (in the noise-free case). The algorithm shows particular promise for discrete optimisation problems where each fitness evaluation is expensive.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.