Brain-inspired learning in physical hardware has enormous potential to learn fast at minimal energy expenditure. One of the characteristics of biological learning systems is their ability to learn in the presence of various noise sources. Inspired by this observation, we introduce a novel noise-based learning approach for physical systems implementing multi-layer neural networks. Simulation results show that our approach allows for effective learning whose performance approaches that of the conventional effective yet energy-costly backpropagation algorithm. Using a spintronics hardware implementation, we demonstrate experimentally that learning can be achieved in a small network composed of physical stochastic magnetic tunnel junctions. These results provide a path towards efficient learning in general physical systems which embraces rather than mitigates the noise inherent in physical devices.

翻译：在物理硬件中实现类脑学习具有以最小能耗快速学习的巨大潜力。生物学习系统的特征之一是在各种噪声源存在的情况下仍能学习的能力。受此启发，我们提出了一种新颖的基于噪声的学习方法，用于实现多层神经网络的物理系统。仿真结果表明，我们的方法能够实现有效的学习，其性能接近传统有效但能耗高昂的反向传播算法。通过自旋电子硬件实现，我们通过实验证明，在由物理随机磁性隧道结组成的小型网络中可以实现学习。这些结果为通用物理系统中的高效学习提供了一条路径——该方法并非抑制物理器件固有的噪声，而是充分利用噪声特性。