StochEP: Stochastic Equilibrium Propagation for Spiking Convergent Recurrent Neural Networks
This work addresses the problem of scalable and biologically plausible training for spiking neural networks, which is incremental by building on existing EP frameworks with a stochastic adaptation.
The paper tackled the challenge of training spiking neural networks (SNNs) with biologically plausible methods by proposing a stochastic equilibrium propagation (EP) framework that integrates probabilistic spiking neurons, which smoothens the optimization landscape and stabilizes training. This approach narrows the performance gap to backpropagation-trained SNNs and EP-trained non-spiking networks in vision benchmarks while preserving locality.
Spiking Neural Networks (SNNs) promise energy-efficient, sparse, biologically inspired computation. Training them with Backpropagation Through Time (BPTT) and surrogate gradients achieves strong performance but remains biologically implausible. Equilibrium Propagation (EP) provides a more local and biologically grounded alternative. However, existing EP frameworks, primarily based on deterministic neurons, either require complex mechanisms to handle discontinuities in spiking dynamics or fail to scale beyond simple visual tasks. Inspired by the stochastic nature of biological spiking mechanism and recent hardware trends, we propose a stochastic EP framework that integrates probabilistic spiking neurons into the EP paradigm. This formulation smoothens the optimization landscape, stabilizes training, and enables scalable learning in deep convolutional spiking convergent recurrent neural networks (CRNNs). We provide theoretical guarantees showing that the proposed stochastic EP dynamics approximate deterministic EP under mean-field theory, thereby inheriting its underlying theoretical guarantees. The proposed framework narrows the gap to both BPTT-trained SNNs and EP-trained non-spiking CRNNs in vision benchmarks while preserving locality, highlighting stochastic EP as a promising direction for neuromorphic and on-chip learning.