Reinforcement learning enhanced quantum-inspired algorithm for combinatorial optimization

Quantum hardware and quantum-inspired algorithms are becoming increasingly popular for combinatorial optimization. However, these algorithms may require careful hyperparameter tuning for each problem instance. We use a reinforcement learning agent in conjunction with a quantum-inspired algorithm to...

Full description

Saved in:
Bibliographic Details
Published in:Machine learning: science and technology Vol. 2; no. 2; pp. 25009 - 25020
Main Authors: Beloborodov, Dmitrii, Ulanov, A E, Foerster, Jakob N, Whiteson, Shimon, Lvovsky, A I
Format: Journal Article
Language:English
Published: Bristol IOP Publishing 01.06.2021
Subjects:
Algorithms
Combinatorial analysis
combinatorial optimization
Ising model
Machine learning
maximum cut
Optimization
reinforcement learning
Training
Tuning
ISSN:2632-2153, 2632-2153
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Quantum hardware and quantum-inspired algorithms are becoming increasingly popular for combinatorial optimization. However, these algorithms may require careful hyperparameter tuning for each problem instance. We use a reinforcement learning agent in conjunction with a quantum-inspired algorithm to solve the Ising energy minimization problem, which is equivalent to the Maximum Cut problem. The agent controls the algorithm by tuning one of its parameters with the goal of improving recently seen solutions. We propose a new Rescaled Ranked Reward (R3) method that enables a stable single-player version of self-play training and helps the agent escape local optima. The training on any problem instance can be accelerated by applying transfer learning from an agent trained on randomly generated problems. Our approach allows sampling high quality solutions to the Ising problem with high probability and outperforms both baseline heuristics and a black-box hyperparameter optimization approach.
Bibliography:MLST-100171.R2
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2632-2153
2632-2153
DOI:10.1088/2632-2153/abc328