RecLight: A Recurrent Neural Network Accelerator with Integrated Silicon Photonics. (arXiv:2209.00084v1 [cs.LG])

Recurrent Neural Networks (RNNs) are used in applications that learn
dependencies in data sequences, such as speech recognition, human activity
recognition, and anomaly detection. In recent years, newer RNN variants, such
as GRUs and LSTMs, have been used for implementing these applications. As many
of these applications are employed in real-time scenarios, accelerating
RNN/LSTM/GRU inference is crucial. In this paper, we propose a novel photonic
hardware accelerator called RecLight for accelerating simple RNNs, GRUs, and
LSTMs. Simulation results indicate that …

arxiv network neural network photonics recurrent neural network silicon silicon photonics