| タイトル |
Amoeba-inspired SAT Solver |
| 著者名 |
Masashi Aono, Song-Ju Kim, Liping Zhu, Makoto Naruse, Motoichi Ohtsu, Hirokazu Hori, Masahiko Hara, |
Vo.No.
開始ページ.終了ページ |
Vol.1, No.586, pp.586-589 |
| 種別 |
|
| 専門 |
|
| キーワード |
|
| 公開日 |
|
| 要旨 |
We propose a biologically-inspired computing algorithm called “AmoebaSAT” for solving an NP-complete combinatorial optimization problem, the Boolean satisfiability problem (SAT). AmoebaSAT is a hybrid of two dynamics; chaotic oscillatory dynamics for exploring the state space are combined with spatiotemporal control dynamics for bouncing back logically-false state transitions. For the former, we employ the logistic map as a unit for generating chaotic fluctuation. The control principle of the latter that we call “bounceback control” is designed to stabilize a state only when it represents a solution, i.e., a satisfiable assignment. We show that, for some benchmark problem instances, AmoebaSAT finds a solution faster than a well-known algorithm called “WalkSAT”, which is considered to be one of the fastest algorithms.
|
| DOI |
10.15248/proc.1.586
|
| PDF |
[download]PDFをダウンロード (310KB)
|