无需反向传播，局部突触规则可实现自监督学习梯度。
Local Synaptic Rules Can Implement a SIGReg Gradient Without Backpropagation
- 结合STDP+与稳态可塑性，仅用局部信号实现精确梯度。
- 时序有序输入使聚类分离度提升至2.49，随机输入仅0.83。
- 在时序MNIST上达87.3%准确率，适合研究神经启发学习机制者。
我们证明，两种经典的局部突触学习规则——尖峰时间依赖可塑性（STDP⁺）的增强部分与稳态可塑性（此处以闪光粒细胞类神经元实现）——共同可实现类似SIGReg的自监督学习目标的精确梯度。该等价性无需梯度计算、全局误差信号、权重传输或标签信息，仅需前/后突触发放率、局部发放统计及自然感官流的时间连续性作为输入。在一项合成聚类任务中，有序呈现使聚类分离度（CSR）升至2.49，而随机排列仅维持基线水平（0.83），约三倍（≈3.5σ）差异完全归因于输入时序。在时序MNIST上，仅使用这些规则训练的两层网络实现了87.3%的线性探测准确率，表明该机制可端到端运行。
原文摘要 · Abstract (English)
We prove that two canonical local synaptic learning rules, the potentiation arm of spike-timing-dependent plasticity (STDP$^+$) and homeostatic plasticity (instantiated here via flashlight granule-cell-like neurons), together can implement the exact gradient of a SIGReg-like self-supervised learning objective. The equivalence requires no gradient calculations, no global error signals, no weight transport, and no label information: the only inputs are pre- and post-synaptic firing rates, local firing statistics, and the temporal contiguity of natural sensory streams. On a synthetic clustering task designed to probe whether class structure can be recovered from temporal ordering of inputs alone, ordered presentation raised cluster separation (CSR) to 2.49 while random ordering left it near baseline (0.83), a roughly threefold ($\approx 3.5σ$) separation attributable solely to input ordering. On temporally ordered MNIST, a two-layer network trained entirely with these rules achieved 87.3% linear-probe accuracy, showing that the mechanism functions end-to-end.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。