Switchable Lightweight Anti-symmetric Processing (SLAP) with CNN Outspeeds Data Augmentation by Smaller Sample -- Application in Gomoku Reinforcement Learning

TL;DR

This paper introduces SLAP, a novel, model-independent method that enhances learning speed and reduces sample requirements in CNNs and reinforcement learning, demonstrated through Gomoku game experiments.

Contribution

SLAP is a new protocol that produces consistent outputs across transformations, significantly speeding up CNN training and reducing sample needs without data augmentation.

Findings

SLAP improved CNN convergence speed by 83%.

SLAP reduced training samples by a factor of 8 in Gomoku reinforcement learning.

SLAP achieved similar performance to data augmentation in reinforcement learning.

Abstract

To replace data augmentation, this paper proposed a method called SLAP to intensify experience to speed up machine learning and reduce the sample size. SLAP is a model-independent protocol/function to produce the same output given different transformation variants. SLAP improved the convergence speed of convolutional neural network learning by 83% in the experiments with Gomoku game states, with only one eighth of the sample size compared with data augmentation. In reinforcement learning for Gomoku, using AlphaGo Zero/AlphaZero algorithm with data augmentation as baseline, SLAP reduced the number of training samples by a factor of 8 and achieved similar winning rate against the same evaluator, but it was not yet evident that it could speed up reinforcement learning. The benefits should at least apply to domains that are invariant to symmetry or certain transformations. As future work, SLAP may aid more explainable learning and transfer learning for domains that are not invariant to symmetry, as a small step towards artificial general intelligence.