利用自旋电子学器件中的噪声实现低能耗神经网络学习
Noise-based Local Learning using Stochastic Magnetic Tunnel Junctions
- 用随机磁隧道结的固有噪声替代传统反向传播
- 实验验证小规模网络可成功学习，性能接近传统算法
- 适合追求低功耗、类脑硬件加速的研究者
类脑学习在物理硬件中具有快速、低功耗的巨大潜力。生物学习系统的一个特征是能在多种噪声环境下学习。受此启发，我们提出一种新型基于噪声的物理学习方法，适用于实现多层神经网络的物理系统。仿真结果表明，该方法可实现高效学习，性能接近传统但高能耗的反向传播算法。通过自旋电子学硬件实现，我们实验验证了由物理随机磁隧道结组成的小型网络可完成学习。这些结果为普遍性物理系统中拥抱而非抑制固有噪声的高效学习提供了路径。
原文摘要 · Abstract (English)
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.
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。