As neural networks grow larger and more complex and data-hungry, training costs are skyrocketing. Especially when lifelong learning is necessary, such as in recommender systems or self-driving cars, this might soon become unsustainable. In this study, we present the first optical co-processor able to accelerate the training phase of digitally-implemented neural networks. We rely on direct feedback alignment as an alternative to backpropagation, and perform the error projection step optically. Leveraging the optical random projections delivered by our co-processor, we demonstrate its use to train a neural network for handwritten digits recognition.

中文翻译：

---

Light-in-the-loop：使用光子协处理器进行神经网络的可扩展训练

随着神经网络变得越来越大、越来越复杂和需要大量数据，训练成本也在飞涨。尤其是在需要终身学习的情况下，例如在推荐系统或自动驾驶汽车中，这可能很快就会变得不可持续。在这项研究中，我们展示了第一个能够加速数字实现神经网络训练阶段的光学协处理器。我们依靠直接反馈对齐作为反向传播的替代方法，并以光学方式执行误差投影步骤。利用我们的协处理器提供的光学随机投影，我们展示了它用于训练用于手写数字识别的神经网络。