用物理振荡器实现快速神经网络训练，无需反向传播。
Learning at the Speed of Physics: Equilibrium Propagation on Oscillator Ising Machines
- 将能量下降原理融入振荡器系统，实现局部学习规则。
- 在MNIST上达97.2%准确率，对硬件噪声鲁棒。
- 适合追求高速低功耗的类脑计算应用。
自然进行能量下降的物理系统为加速机器学习提供了直接路径。振荡器伊辛机（OIMs）体现了这一思想：其千兆赫频率动态既模拟了能量模型（EBMs）的优化过程，也对应损失曲面的梯度下降，而内在噪声则对应朗之万动力学，支持采样与优化并行。平衡传播（EP）将这些过程统一到单一总能量曲面上，实现无需全局反向传播的局部学习规则。我们证明，在OIM上实施的EP在MNIST上达到约97.2±0.1%的准确率，在Fashion-MNIST上达到约88.0±0.1%，同时在参数量化和相位噪声等真实硬件约束下仍保持稳健。这些结果确立了OIM作为高速、低功耗类脑学习的物理载体，表明常受传统处理器瓶颈限制的能量模型，可能在直接执行其优化的物理硬件上实现实际应用。
原文摘要 · Abstract (English)
Physical systems that naturally perform energy descent offer a direct route to accelerating machine learning. Oscillator Ising Machines (OIMs) exemplify this idea: their GHz-frequency dynamics mirror both the optimization of energy-based models (EBMs) and gradient descent on loss landscapes, while intrinsic noise corresponds to Langevin dynamics - supporting sampling as well as optimization. Equilibrium Propagation (EP) unifies these processes into descent on a single total energy landscape, enabling local learning rules without global backpropagation. We show that EP on OIMs achieves competitive accuracy ($\sim 97.2 \pm 0.1 \%$ on MNIST, $\sim 88.0 \pm 0.1 \%$ on Fashion-MNIST), while maintaining robustness under realistic hardware constraints such as parameter quantization and phase noise. These results establish OIMs as a fast, energy-efficient substrate for neuromorphic learning, and suggest that EBMs - often bottlenecked by conventional processors - may find practical realization on physical hardware whose dynamics directly perform their optimization.
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。