Proceedings ArticleDOI
O2NN: Optical Neural Networks with Differential Detection-Enabled Optical Operands
Jiaqi Gu,Zheng Zhao,Chenghao Feng,Zhoufeng Ying,Ray T. Chen,David Z. Pan +5 more
- pp 1062-1067
Reads0
Chats0
TLDR
Wang et al. as mentioned in this paper proposed a novel ONN engine O2NN based on wavelength-division multiplexing and differential detection to enable high-performance, robust, and versatile photonic neural computing with both light operands.Abstract:
Optical neuromorphic computing has demonstrated promising performance with ultra-high computation speed, high bandwidth, and low energy consumption. The traditional optical neural network (ONN) architectures realize neuromorphic computing via electrical weight encoding. However, previous ONN design methodologies can only handle static linear projection with stationary synaptic weights, thus fail to support efficient and flexible computing when both operands are dynamically-encoded light signals. In this work, we propose a novel ONN engine O2NN based on wavelength-division multiplexing and differential detection to enable high-performance, robust, and versatile photonic neural computing with both light operands. Balanced optical weights and augmented quantization are introduced to enhance the representability and efficiency of our architecture. Static and dynamic variations are discussed in detail with a knowledge-distillation-based solution given for robustness improvement. Discussions on hardware cost and efficiency are provided for a comprehensive comparison with prior work. Simulation and experimental results show that the proposed ONN architecture provides flexible, efficient, and robust support for high-performance photonic neural computing with fully-optical operands under low-bit quantization and practical variations.read more
Citations
More filters
Proceedings ArticleDOI
ELight: Enabling Efficient Photonic In-Memory Neurocomputing with Life Enhancement
TL;DR: In this paper , a synergistic optimization framework, ELight, is proposed to minimize the overall write efforts for efficient and reliable optical in-memory neurocomputing, which reduces the total number of writes and dynamic power with comparable accuracy.
Journal ArticleDOI
Light in AI: Toward Efficient Neurocomputing With Optical Neural Networks—A Tutorial
TL;DR: An overview of state-of-the-art cross-layer co-design methodologies for scalable, robust, and self-learnable ONN designs across the circuit, architecture, and algorithm levels is given.
Posted Content
Towards Memory-Efficient Neural Networks via Multi-Level in situ Generation
TL;DR: In this article, a general and unified framework is proposed to trade expensive memory transactions with ultra-fast on-chip computations, directly translating to performance improvement by jointly exploring the intrinsic correlations and bit-level redundancy within DNN kernels and propose a multi-level in situ generation mechanism with mixed-precision bases.
Journal ArticleDOI
Silicon Photonics for Future Computing Systems
TL;DR: In this paper , the authors provide an overview of silicon photonics technology and its applications in the design and improvement of current and future computing systems, and discuss several research opportunities to push forward the application of silicon-on-insulator (SOI) waveguides.
Journal ArticleDOI
ELight: Toward Efficient and Aging-Resilient Photonic In-Memory Neurocomputing
TL;DR: This work proposes a holistic solution, ELight, to tackle both the aging issue and the post-aging reliability issue, where a proactive aging- aware optimization framework minimizes the overall PCM write cost and a post- aging tolerance scheme overcomes the effect of aged PCM.
References
More filters
Proceedings Article
Attention is All you Need
Ashish Vaswani,Noam Shazeer,Niki Parmar,Jakob Uszkoreit,Llion Jones,Aidan N. Gomez,Lukasz Kaiser,Illia Polosukhin +7 more
TL;DR: This paper proposed a simple network architecture based solely on an attention mechanism, dispensing with recurrence and convolutions entirely and achieved state-of-the-art performance on English-to-French translation.
Journal ArticleDOI
Deep learning with coherent nanophotonic circuits
TL;DR: A new architecture for a fully optical neural network is demonstrated that enables a computational speed enhancement of at least two orders of magnitude and three order of magnitude in power efficiency over state-of-the-art electronics.
Posted Content
DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradients
TL;DR: DoReFa-Net, a method to train convolutional neural networks that have low bitwidth weights and activations using low bit width parameter gradients, is proposed and can achieve comparable prediction accuracy as 32-bit counterparts.
Journal ArticleDOI
Single-chip microprocessor that communicates directly using light
Chen Sun,Chen Sun,Mark T. Wade,Yunsup Lee,Jason S. Orcutt,Jason S. Orcutt,Luca Alloatti,Michael Georgas,Andrew Waterman,Jeffrey M. Shainline,Jeffrey M. Shainline,Rimas Avizienis,Sen Lin,Benjamin Moss,Rajesh Kumar,Fabio Pavanello,Amir H. Atabaki,Henry Cook,Albert Ou,Jonathan Leu,Yu-Hsin Chen,Krste Asanovic,Rajeev J. Ram,Milos A. Popovic,Vladimir Stojanovic +24 more
TL;DR: This demonstration could represent the beginning of an era of chip-scale electronic–photonic systems with the potential to transform computing system architectures, enabling more powerful computers, from network infrastructure to data centres and supercomputers.
Journal ArticleDOI
Neuromorphic photonic networks using silicon photonic weight banks.
Alexander N. Tait,Thomas Ferreira de Lima,Ellen Zhou,Allie X. Wu,Mitchell A. Nahmias,Bhavin J. Shastri,Paul R. Prucnal +6 more
TL;DR: First observations of a recurrent silicon photonic neural network, in which connections are configured by microring weight banks are reported, and a mathematical isomorphism between the silicon photonics circuit and a continuous neural network model is demonstrated through dynamical bifurcation analysis.