CaRe-BN: Precise Moving Statistics for Stabilizing Spiking Neural Networks in Reinforcement Learning

• URL: http://arxiv.org/abs/2509.23791v1
• Date: Sun, 28 Sep 2025 10:21:17 GMT
• Title: CaRe-BN: Precise Moving Statistics for Stabilizing Spiking Neural Networks in Reinforcement Learning
• Authors: Zijie Xu, Xinyu Shi, Yiting Dong, Zihan Huang, Zhaofei Yu,
• Abstract summary: Spiking Neural Networks (SNNs) offer low-latency and energy-efficient decision-making on neuromorphic hardware.<n>Due to the discrete and non-differentiable nature of spikes, directly trained SNNs rely heavily on Batch Normalization (BN) to stabilize gradient updates.<n>In online Reinforcement Learning (RL), BN statistics hinder exploitation, resulting in slower convergence and suboptimal policies.
• Score: 50.87795054453648
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Spiking Neural Networks (SNNs) offer low-latency and energy-efficient decision-making on neuromorphic hardware by mimicking the event-driven dynamics of biological neurons. However, due to the discrete and non-differentiable nature of spikes, directly trained SNNs rely heavily on Batch Normalization (BN) to stabilize gradient updates. In online Reinforcement Learning (RL), imprecise BN statistics hinder exploitation, resulting in slower convergence and suboptimal policies. This challenge limits the adoption of SNNs for energy-efficient control on resource-constrained devices. To overcome this, we propose Confidence-adaptive and Re-calibration Batch Normalization (CaRe-BN), which introduces (\emph{i}) a confidence-guided adaptive update strategy for BN statistics and (\emph{ii}) a re-calibration mechanism to align distributions. By providing more accurate normalization, CaRe-BN stabilizes SNN optimization without disrupting the RL training process. Importantly, CaRe-BN does not alter inference, thus preserving the energy efficiency of SNNs in deployment. Extensive experiments on continuous control benchmarks demonstrate that CaRe-BN improves SNN performance by up to $22.6\%$ across different spiking neuron models and RL algorithms. Remarkably, SNNs equipped with CaRe-BN even surpass their ANN counterparts by $5.9\%$. These results highlight a new direction for BN techniques tailored to RL, paving the way for neuromorphic agents that are both efficient and high-performing.

Related papers

• S$^2$NN: Sub-bit Spiking Neural Networks [53.08060832135342]
  Spiking Neural Networks (SNNs) offer an energy-efficient paradigm for machine intelligence.<n>Despite recent advances in binary SNNs, the storage and computational demands remain substantial for large-scale networks.<n>We propose Sub-bit Spiking Neural Networks (S$2$NNs) that represent weights with less than one bit.
  arXiv  Detail & Related papers  (2025-09-29T04:17:44Z)
• A Self-Ensemble Inspired Approach for Effective Training of Binary-Weight Spiking Neural Networks [66.80058515743468]
  Training Spiking Neural Networks (SNNs) and Binary Neural Networks (BNNs) is challenging because of the non-differentiable spike generation function.<n>We present a novel perspective on the dynamics of SNNs and their close connection to BNNs through an analysis of the backpropagation process.<n>Specifically, we leverage a structure of multiple shortcuts and a knowledge distillation-based training technique to improve the training of (binary-weight) SNNs.
  arXiv  Detail & Related papers  (2025-08-18T04:11:06Z)
• Proxy Target: Bridging the Gap Between Discrete Spiking Neural Networks and Continuous Control [26.105497272647977]
  Spiking Neural Networks (SNNs) offer low-latency and energy-efficient decision making through neuromorphic hardware.<n>Recent studies overlook whether Reinforcement Learning (RL) algorithms are suitable for SNNs.<n>We propose a novel proxy target framework to bridge the gap between discrete SNN and continuous control.
  arXiv  Detail & Related papers  (2025-05-30T03:08:03Z)
• Towards Accurate Binary Spiking Neural Networks: Learning with Adaptive Gradient Modulation Mechanism [14.425611637823511]
  Binary Spiking Neural Networks (BSNNs) inherit the eventdriven paradigm of SNNs, while also adopting the reduced storage burden of binarization techniques.<n>These distinct advantages grant BSNNs lightweight and energy-efficient characteristics, rendering them ideal for deployment on resource-constrained edge devices.<n>However, due to the binary synaptic weights and non-differentiable spike function, effectively training BSNNs remains an open question.
  arXiv  Detail & Related papers  (2025-02-20T07:59:08Z)
• BN-SCAFFOLD: controlling the drift of Batch Normalization statistics in Federated Learning [2.563180814294141]
  Federated Learning (FL) is gaining traction as a learning paradigm for training Machine Learning (ML) models in a decentralized way. Batch Normalization (BN) is ubiquitous in Deep Neural Networks (DNN) BN has been reported to hinder performance of DNNs in heterogeneous FL. We introduce a unified theoretical framework for analyzing the convergence of variance reduction algorithms in the BN-DNN setting.
  arXiv  Detail & Related papers  (2024-10-04T09:53:20Z)
• Converting High-Performance and Low-Latency SNNs through Explicit Modelling of Residual Error in ANNs [27.46147049872907]
  Spiking neural networks (SNNs) have garnered interest due to their energy efficiency and superior effectiveness on neuromorphic chips. One of the mainstream approaches to implementing deep SNNs is the ANN-SNN conversion. We propose a new approach based on explicit modeling of residual errors as additive noise.
  arXiv  Detail & Related papers  (2024-04-26T14:50:46Z)
• STCSNN: High energy efficiency spike-train level spiking neural networks with spatio-temporal conversion [4.892303151981707]
  Brain-temporal spiking neuron networks (SNNs) have attracted widespread research interest due to their low power features, high biological plausibility, and strong information processing capability. Although adopting a surrogate (SG) makes the non-differentiability SNN trainable, achieving comparable accuracy for ANNs and keeping low-power features simultaneously is still tricky. In this paper, we proposed an energy-efficient spiking neural network withtemporal conversion, which has low computational cost and high accuracy.
  arXiv  Detail & Related papers  (2023-07-14T03:27:34Z)
• Skip Connections in Spiking Neural Networks: An Analysis of Their Effect on Network Training [0.8602553195689513]
  Spiking neural networks (SNNs) have gained attention as a promising alternative to traditional artificial neural networks (ANNs) In this paper, we study the impact of skip connections on SNNs and propose a hyper parameter optimization technique that adapts models from ANN to SNN. We demonstrate that optimizing the position, type, and number of skip connections can significantly improve the accuracy and efficiency of SNNs.
  arXiv  Detail & Related papers  (2023-03-23T07:57:32Z)
• Spatial-Temporal-Fusion BNN: Variational Bayesian Feature Layer [77.78479877473899]
  We design a spatial-temporal-fusion BNN for efficiently scaling BNNs to large models. Compared to vanilla BNNs, our approach can greatly reduce the training time and the number of parameters, which contributes to scale BNNs efficiently.
  arXiv  Detail & Related papers  (2021-12-12T17:13:14Z)
• "BNN - BN = ?": Training Binary Neural Networks without Batch Normalization [92.23297927690149]
  Batch normalization (BN) is a key facilitator and considered essential for state-of-the-art binary neural networks (BNN) We extend their framework to training BNNs, and for the first time demonstrate that BNs can be completed removed from BNN training and inference regimes.
  arXiv  Detail & Related papers  (2021-04-16T16:46:57Z)
• Batch Normalization Increases Adversarial Vulnerability and Decreases Adversarial Transferability: A Non-Robust Feature Perspective [91.5105021619887]
  Batch normalization (BN) has been widely used in modern deep neural networks (DNNs) BN is observed to increase the model accuracy while at the cost of adversarial robustness. It remains unclear whether BN mainly favors learning robust features (RFs) or non-robust features (NRFs)
  arXiv  Detail & Related papers  (2020-10-07T10:24:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.