Open AccessPosted Content
CrossLight: A Cross-Layer Optimized Silicon Photonic Neural Network Accelerator
Reads0
Chats0
TLDR
CrossLight as discussed by the authors proposes a cross-layer optimized neural network accelerator called CrossLight that leverages silicon photonics, which 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.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.read more
Citations
More filters
Posted Content
SONIC: A Sparse Neural Network Inference Accelerator with Silicon Photonics for Energy-Efficient Deep Learning.
TL;DR: SONIC as discussed by the authors proposes a novel silicon photonics-based sparse neural network inference accelerator, which can achieve up to 5.8x better performance per watt and 8.4x lower energy-per-bit than state-of-the-art sparse electronic neural network accelerators.
Posted Content
O-HAS: Optical Hardware Accelerator Search for Boosting Both Acceleration Performance and Development Speed.
TL;DR: In this paper, the authors developed a framework called O-HAS, which can automatically explore the large design space of optical DNN accelerators and identify the optimal accelerators (i.e., the micro-architectures and algorithm-to-accelerator mapping methods).
Posted Content
An Electro-Photonic System for Accelerating Deep Neural Networks
Cansu Demirkiran,Furkan Eris,Gongyu Wang,Jonathan Elmhurst,Nick Moore,Nicholas C. Harris,Ayon Basumallik,Vijay Janapa Reddi,Ajay Joshi,Darius Bunandar +9 more
TL;DR: In this article, a hybrid electro-photonic system is proposed to accelerate DNNs, which includes an electronic host processor and DRAM, and a custom electrophotonic hardware accelerator called ADEPT.
References
More filters
Journal ArticleDOI
Gradient-based learning applied to document recognition
Yann LeCun,Léon Bottou,Léon Bottou,Yoshua Bengio,Yoshua Bengio,Yoshua Bengio,Patrick Haffner +6 more
TL;DR: In this article, a graph transformer network (GTN) is proposed for handwritten character recognition, which can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters.
Proceedings ArticleDOI
In-Datacenter Performance Analysis of a Tensor Processing Unit
Norman P. Jouppi,Cliff Young,Nishant Patil,David A. Patterson,Gaurav Agrawal,Raminder Bajwa,Sarah Bates,Suresh Bhatia,Nan Boden,Albert T. Borchers,Rick Boyle,Pierre-luc Cantin,Clifford Chao,Christopher Aaron Clark,Jeremy Coriell,Michael J. Daley,Matt Dau,Jeffrey Dean,Ben Gelb,Tara Vazir Ghaemmaghami,Rajendra Gottipati,William John Gulland,Robert Hagmann,C. Richard Ho,Doug Hogberg,John Hu,Robert Hundt,D. Hurt,Julian Ibarz,Aaron Jaffey,Alek Jaworski,Alexander Kaplan,Khaitan Harshit,Daniel Killebrew,Andy Koch,Naveen Kumar,Steve Lacy,James Laudon,James Law,Diemthu Le,Chris Leary,Zhuyuan Liu,Kyle Lucke,Alan Lundin,Gordon MacKean,Adriana Maggiore,Maire Mahony,Kieran Miller,Rahul Nagarajan,Ravi Narayanaswami,Ray Ni,Kathy Nix,Thomas Norrie,Mark Omernick,Narayana Penukonda,Andrew Everett Phelps,Jonathan Ross,Matt Ross,Amir Salek,Emad Samadiani,Chris Severn,Gregory Sizikov,Matthew Snelham,Jed Souter,Dan Steinberg,Andy Swing,Mercedes Tan,Gregory Michael Thorson,Bo Tian,Horia Toma,Erick Tuttle,Vijay K. Vasudevan,Richard Walter,Walter Wang,Eric Wilcox,Doe Hyun Yoon +75 more
TL;DR: The Tensor Processing Unit (TPU) as discussed by the authors is a custom ASIC deployed in datacenters since 2015 that accelerates the inference phase of neural networks (NN) using a 65,536 8-bit MAC matrix multiply unit that offers a peak throughput of 92 TeraOps/second (TOPS).
Journal ArticleDOI
Silicon microring resonators
Wim Bogaerts,P. De Heyn,T. Van Vaerenbergh,K. De Vos,S. Kumar Selvaraja,Tom Claes,Pieter Dumon,Peter Bienstman,D. Van Thourhout,Roel Baets +9 more
TL;DR: An overview of the current state-of-the-art in silicon nanophotonic ring resonators is presented in this paper, where the basic theory of ring resonance is discussed and applied to the peculiarities of submicron silicon photonic wire waveguides: the small dimensions and tight bend radii, sensitivity to perturbations and the boundary conditions of the fabrication processes.
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.
Journal ArticleDOI
The chips are down for Moore's law.
TL;DR: The semiconductor industry will soon abandon its pursuit of Moore's law as mentioned in this paper, and things could get a lot more interesting in the next few decades, which is a good thing.