CrossLight: A Cross-Layer Optimized Silicon Photonic Neural Network Accelerator

2021/02/13 by Sunny, Febin, Mirza, Asif, Nikdast, Mahdi +1 · 2 citations
#Emerging Technologies (cs.ET) #FOS: Computer and information sciences #Hardware Architecture (cs.AR) #Machine Learning (cs.LG) #Neural and Evolutionary Computing (cs.NE)

paper · doi:10.48550/arxiv.2102.06960

Abstract

Domain-specific neural network accelerators have seen growing interest in recent years due to their improved energy efficiency and inference performance compared to CPUs and GPUs. In this paper, we propose a novel cross-layer optimized neural network accelerator called CrossLight that leverages silicon photonics. CrossLight includes device-level engineering for resilience to process variations and thermal crosstalk, circuit-level tuning enhancements for inference latency reduction, and architecture-level optimization to enable higher resolution, better energy-efficiency, and improved throughput. On average, CrossLight offers 9.5x lower energy-per-bit and 15.9x higher performance-per-watt at 16-bit resolution than state-of-the-art photonic deep learning accelerators.

Cited by

• Opto-ViT: Architecting a Near-Sensor Region of Interest-Aware Vision Transformer Accelerator with Silicon Photonics
• In situ fine-tuning of in silico trained Optical Neural Networks

Related

• A Survey on Silicon Photonics for Deep Learning
• SONIC: A Sparse Neural Network Inference Accelerator with Silicon Photonics for Energy-Efficient Deep Learning
• Weight Bank Addition Photonic Accelerator for Artificial Intelligence
• A Winograd-based Integrated Photonics Accelerator for Convolutional Neural Networks
• BPLight-CNN: A Photonics-based Backpropagation Accelerator for Deep Learning