A Gradient-based yet Spike-Timing-Dependent Solution to the Feedback Learning Problem in Neural Microcircuits
Abstract
The brain uses discrete spikes for dynamic computation, yet, how neural microcircuits (NMCs) solve temporal credit assignment using local spike timing remains a fundamental open question. Dominant spiking neural network (SNN) approaches circumvent this by approximating backpropagation through surrogate gradients, decoupling learning from biological spike timing. Here, we reformulate temporal credit assignment as a state separation problem: extracting task-required components induced by historical perturbations directly from the current neural state. This enables an online feedback learning framework for NMCs through a gradient tunneling (GT) algorithm and the lead-lag expansion technique that derives credit assignment from local synaptic spike timing, while remaining compatible with ANN-SNN hybrid architectures. Experimentally, GT-trained NMCs excel at long-timescale evidence integration and noise-robust memory retention, and perform comparably to leading SNN online learning methods on real-world benchmarks with far fewer parameters. The proposed framework addresses the two-decade-old NMC feedback learning problem and suggests a computationally plausible explanation for the brain's learning mechanisms.
Related
- Preserving spiking and EEG in detailed models of human cortical microcircuits with morphological reductions
- Striatal interneuron microcircuits gate reinforcement to stabilize adaptive choice
- Closed-loop EEG feedback pacing alters task load during digital learning
- Are cortical microcircuits optimized for information flux?—A simulation-based reverse engineering study
- Feedback reliability modulates learning-related dynamics of midfrontal error monitoring
BibTeX
@misc{zhang2026gradientbased,
title = {{A Gradient-based yet Spike-Timing-Dependent Solution to the Feedback Learning Problem in Neural Microcircuits}},
author = {Xiangnan Zhang and Jingxin Liu and Ranqi Lu and Jingyu Liu and Qunxi Dong and Fuze Tian and Lixian Zhu and Bin Hu and Björn W. Schuller},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.08070},
eprint = {2609.08070},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.08070},
}
RIS
TY - UNPB
TI - A Gradient-based yet Spike-Timing-Dependent Solution to the Feedback Learning Problem in Neural Microcircuits
AU - Xiangnan Zhang
AU - Jingxin Liu
AU - Ranqi Lu
AU - Jingyu Liu
AU - Qunxi Dong
AU - Fuze Tian
AU - Lixian Zhu
AU - Bin Hu
AU - Björn W. Schuller
PY - 2026
DA - 2026/09/10
PB - arXiv
DO - 10.48550/arxiv.2609.08070
UR - https://arxiv.org/abs/2609.08070
AB - The brain uses discrete spikes for dynamic computation, yet, how neural microcircuits (NMCs) solve temporal credit assignment using local spike timing remains a fundamental open question. Dominant spiking neural network (SNN) approaches circumvent this by approximating backpropagation through surrogate gradients, decoupling learning from biological spike timing. Here, we reformulate temporal credit assignment as a state separation problem: extracting task-required components induced by historical perturbations directly from the current neural state. This enables an online feedback learning framework for NMCs through a gradient tunneling (GT) algorithm and the lead-lag expansion technique that derives credit assignment from local synaptic spike timing, while remaining compatible with ANN-SNN hybrid architectures. Experimentally, GT-trained NMCs excel at long-timescale evidence integration and noise-robust memory retention, and perform comparably to leading SNN online learning methods on real-world benchmarks with far fewer parameters. The proposed framework addresses the two-decade-old NMC feedback learning problem and suggests a computationally plausible explanation for the brain's learning mechanisms.
ER -