no code implementations • 31 Dec 2023 • Yequan Zhao, Xian Xiao, Xinling Yu, Ziyue Liu, Zhixiong Chen, Geza Kurczveil, Raymond G. Beausoleil, Zheng Zhang
Despite the ultra-high speed of optical neural networks, training a PINN on an optical chip is hard due to (1) the large size of photonic devices, and (2) the lack of scalable optical memory devices to store the intermediate results of back-propagation (BP).
no code implementations • 18 Aug 2023 • Yequan Zhao, Xinling Yu, Zhixiong Chen, Ziyue Liu, Sijia Liu, Zheng Zhang
Backward propagation (BP) is widely used to compute the gradients in neural network training.
no code implementations • 17 Feb 2023 • Yequan Zhao, Xian Xiao, Geza Kurczveil, Raymond G. Beausoleil, Zheng Zhang
We propose the first tensorized optical multimodal fusion network architecture with a self-attention mechanism and low-rank tensor fusion.