用自旋机动力学改进能量学习，让神经网络更快更稳地训练。
Hybridizing Equilibrium Propagation with Ising Machines for Efficient Energy-Based Learning
- 引入自旋机相空间动态替代传统松弛，改变神经态达平衡的路径。
- 在MNIST、FashionMNIST、CIFAR-10上达到与反向传播相当的精度。
- 加速收敛、降低能量壁垒，提升噪声鲁棒性，适合低功耗计算场景。
人工智能的快速发展推动了深度神经网络的重大进步。然而，传统基于GPU的训练方式能耗极高，促使人们探索物理动力学与兼容的能量学习方案，如平衡传播（EP）。但EP常因相空间收缩而陷入局部极小值。本文提出一种受自旋机动力学启发的平衡传播框架，将耗散的霍普菲尔德松弛替换为包含共轭变量的扩展相空间动态。该方法保持了EP的局部两阶段学习规则，但改变了神经态达平衡的物理路径。实验表明，这种动力学有效降低了能量壁垒，加速收敛，提升噪声鲁棒性，并在MNIST、FashionMNIST和CIFAR-10上实现了与反向传播相当的性能。
原文摘要 · Abstract (English)
The rapid evolution of artificial intelligence has led to substantial advances in deep neural networks. Nonetheless, conventional GPU-based training remains highly energy-demanding, motivating the exploration of physical dynamics and compatible energy-based learning schemes, such as equilibrium propagation (EP). EP-based training, however, frequently suffers from convergence to local minima due to phase-space contraction. Here we introduce an Ising-dynamics-inspired equilibrium-propagation framework in which dissipative Hopfield relaxation is replaced by an extended phase-space dynamics with conjugate variables. The resulting training paradigm keeps the local two-phase learning rule of EP while changing the physical route by which neural states reach equilibrium. We show that this dynamics lowers effective energy barriers, accelerates convergence, improves noise robustness, and trains deep convolutional Hopfield networks on MNIST, FashionMNIST, and CIFAR-10 with performance comparable to backpropagation.
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。