GRSN: Gated Recurrent Spiking Neurons for POMDPs and MARL

TL;DR

This paper introduces GRSN, a novel spiking neural network architecture with gated neurons and a temporal alignment paradigm, enabling efficient reinforcement learning in POMDPs and MARL with reduced power consumption.

Contribution

The paper presents GRSN, a new spiking neural network model that better captures temporal dynamics and improves RL performance under resource constraints.

Findings

Achieves comparable performance to RNNs in POMDPs and MARL tasks.

Reduces power consumption by approximately 50%.

Effectively leverages temporal dynamics of spiking neurons.

Abstract

Spiking neural networks (SNNs) are widely applied in various fields due to their energy-efficient and fast-inference capabilities. Applying SNNs to reinforcement learning (RL) can significantly reduce the computational resource requirements for agents and improve the algorithm's performance under resource-constrained conditions. However, in current spiking reinforcement learning (SRL) algorithms, the simulation results of multiple time steps can only correspond to a single-step decision in RL. This is quite different from the real temporal dynamics in the brain and also fails to fully exploit the capacity of SNNs to process temporal data. In order to address this temporal mismatch issue and further take advantage of the inherent temporal dynamics of spiking neurons, we propose a novel temporal alignment paradigm (TAP) that leverages the single-step update of spiking neurons to accumulate historical state information in RL and introduces gated units to enhance the memory capacity of spiking neurons. Experimental results show that our method can solve partially observable Markov decision processes (POMDPs) and multi-agent cooperation problems with similar performance as recurrent neural networks (RNNs) but with about 50% power consumption.