Sigmoid-Weighted Linear Units for Neural Network Function Approximation in Reinforcement Learning

TL;DR

This paper introduces SiLU and dSiLU activation functions for neural networks in reinforcement learning, demonstrating their effectiveness in surpassing DQN performance in Atari games and other benchmarks.

Contribution

It proposes novel activation functions for reinforcement learning neural networks and shows they can outperform DQN using simpler on-policy methods.

Findings

Achieved state-of-the-art results in Tetris variants with shallow networks.

Outperformed DQN in Atari 2600 games with deep Sarsa(λ) and SiLU/dSiLU units.

Validated the effectiveness of on-policy learning with these activations.

Abstract

In recent years, neural networks have enjoyed a renaissance as function approximators in reinforcement learning. Two decades after Tesauro's TD-Gammon achieved near top-level human performance in backgammon, the deep reinforcement learning algorithm DQN achieved human-level performance in many Atari 2600 games. The purpose of this study is twofold. First, we propose two activation functions for neural network function approximation in reinforcement learning: the sigmoid-weighted linear unit (SiLU) and its derivative function (dSiLU). The activation of the SiLU is computed by the sigmoid function multiplied by its input. Second, we suggest that the more traditional approach of using on-policy learning with eligibility traces, instead of experience replay, and softmax action selection with simple annealing can be competitive with DQN, without the need for a separate target network. We validate our proposed approach by, first, achieving new state-of-the-art results in both stochastic SZ-Tetris and Tetris with a small 10$\times$10 board, using TD($\lambda$) learning and shallow dSiLU network agents, and, then, by outperforming DQN in the Atari 2600 domain by using a deep Sarsa($\lambda$) agent with SiLU and dSiLU hidden units.