GRSN: Gated Recurrent Spiking Neurons for POMDPs and MARL

arXiv:2404.15597v1 Announce Type: cross
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 …

abstract agents algorithm arxiv capabilities computational cs.ai cs.lg cs.ma cs.ne current energy fields however inference networks neural networks neurons performance reduce reinforcement reinforcement learning requirements s performance spiking neural networks the algorithm type