ChipAI: A scalable chiplet-based accelerator for efficient DNN inference using silicon photonics
Authors:
- Hao ZhangHZHao Zhang
- Haibo ZhangHZHaibo Zhang
- Zhiyi HuangZHZhiyi Huang
- Yawen ChenYCYawen Chen
Open Access
- 26 November 2024
- journal article
- research article
- Published by Elsevier in Journal of Systems Architecture
- Vol. 158
Abstract
No abstract availableKeywords
Funding Information
- University of Adelaide
- Kungliga Tekniska Högskolan
This publication has 38 references indexed in Scilit:
- Towards Efficient On-Chip Communication: A Survey on Silicon Nanophotonics and Optical Networks-on-ChipJournal of Systems Architecture, 2024
- Roadmapping the next generation of silicon photonicsNature Communications, 2024
- An efficient real-time accelerator for high-accuracy DNN-based optical flow estimation in FPGAJournal of Systems Architecture, 2022
- Reconfigurable Network-on-Chip based Convolutional Neural Network AcceleratorJournal of Systems Architecture, 2022
- Response time analysis of parallel tasks on accelerator-based heterogeneous platformsJournal of Systems Architecture, 2022
- Photonic matrix multiplication lights up photonic accelerator and beyondLight: Science & Applications, 2022
- Low-Loss and Broadband Nonvolatile Phase-Change Directional Coupler SwitchesACS Photonics, 2019
- Active microring based tunable optical power splittersOptics Communications, 2016
- Deep learningNature, 2015
- Silicon microring resonatorsLaser & Photonics Reviews, 2011