Silicon microring synapses enable photonic deep learning beyond 9-bit precision
arXiv:2104.01164 · doi:10.1364/OPTICA.446100
Abstract
Deep neural networks (DNN) consist of layers of neurons interconnected by synaptic weights. A high bit-precision in weights is generally required to guarantee high accuracy in many applications. Minimizing error accumulation between layers is also essential when building large-scale networks. Recent demonstrations of photonic neural networks are limited in bit-precision due to crosstalk and the high sensitivity of optical components (e.g., resonators). Here, we experimentally demonstrate a record-high precision of 9 bits with a dithering control scheme for photonic synapses. We then numerically simulated the impact with increased synaptic precision on a wireless signal classification application. This work could help realize the potential of photonic neural networks for many practical, real-world tasks.
7 pages
References in corpus (2)
Cited by in corpus (11)
- High-order tensor flow processing using integrated photonic circuits
- In-memory photonic dot-product engine with electrically programmable weight banks
- Silicon Photonic Architecture for Training Deep Neural Networks with Direct Feedback Alignment
- Broadband physical layer cognitive radio with an integrated photonic processor for blind source separation
- Integrated Photonic Tensor Processing Unit for a Matrix Multiply: a Review
- Photonic neural networks based on integrated silicon microresonators
- Symmetric silicon microring resonator optical crossbar array for accelerated inference and training in deep learning
- Real-Time Blind Photonic Interference Cancellation for mmWave MIMO
- Waveguide-multiplexed photonic matrix-vector multiplication processor using multiport photodetectors
- Programmable 200 GOPS Hopfield-inspired photonic Ising machine
- High-precision programming of large-scale ring resonator circuits with minimal pre-calibration