Ranked Reward: Enabling Self-Play Reinforcement Learning for Combinatorial Optimization
This addresses the challenge of enabling self-play benefits for single-player games, which is incremental for combinatorial optimization problems like bin packing.
The paper tackled the problem of applying self-play reinforcement learning to single-player combinatorial optimization tasks by introducing the Ranked Reward (R2) algorithm, which ranks rewards to create a relative performance metric, and results showed it outperforms Monte Carlo tree search, heuristic algorithms, and integer programming solvers on bin packing problems.
Adversarial self-play in two-player games has delivered impressive results when used with reinforcement learning algorithms that combine deep neural networks and tree search. Algorithms like AlphaZero and Expert Iteration learn tabula-rasa, producing highly informative training data on the fly. However, the self-play training strategy is not directly applicable to single-player games. Recently, several practically important combinatorial optimisation problems, such as the travelling salesman problem and the bin packing problem, have been reformulated as reinforcement learning problems, increasing the importance of enabling the benefits of self-play beyond two-player games. We present the Ranked Reward (R2) algorithm which accomplishes this by ranking the rewards obtained by a single agent over multiple games to create a relative performance metric. Results from applying the R2 algorithm to instances of a two-dimensional and three-dimensional bin packing problems show that it outperforms generic Monte Carlo tree search, heuristic algorithms and integer programming solvers. We also present an analysis of the ranked reward mechanism, in particular, the effects of problem instances with varying difficulty and different ranking thresholds.