Channel Tiling for Improved Performance and Accuracy of Optical Neural Network Accelerators
arXiv:2011.07391
Abstract
Low latency, high throughput inference on Convolution Neural Networks (CNNs) remains a challenge, especially for applications requiring large input or large kernel sizes. 4F optics provides a solution to accelerate CNNs by converting convolutions into Fourier-domain point-wise multiplications that are computationally 'free' in optical domain. However, existing 4F CNN systems suffer from the all-positive sensor readout issue which makes the implementation of a multi-channel, multi-layer CNN not scalable or even impractical. In this paper we propose a simple channel tiling scheme for 4F CNN systems that utilizes the high resolution of 4F system to perform channel summation inherently in optical domain before sensor detection, so the outputs of different channels can be correctly accumulated. Compared to state of the art, channel tiling gives similar accuracy, significantly better robustness to sensing quantization (33\% improvement in required sensing precision) error and noise (10dB reduction in tolerable sensing noise), 0.5X total filters required, 10-50X+ throughput improvement and as much as 3X reduction in required output camera resolution/bandwidth. Not requiring any additional optical hardware, the proposed channel tiling approach addresses an important throughput and precision bottleneck of high-speed, massively-parallel optical 4F computing systems.
arXiv comment11 pages, 8 figures
References in corpus (9)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Neuromorphic Silicon Photonic Networks
- SpaceNet: A Remote Sensing Dataset and Challenge Series
- Reprogrammable Electro-Optic Nonlinear Activation Functions for Optical Neural Networks
- Rapid Generation of Light Beams Carrying Orbital Angular Momentum
- Digital Electronics and Analog Photonics for Convolutional Neural Networks (DEAP-CNNs)
- High sampling rate single-pixel digital holography system employing a DMD and phase-encoded patterns
- On-Chip Optical Convolutional Neural Networks
- High-Speed Phase-Only Spatial Light Modulators with Two-Dimensional Tunable Microcavity Arrays