2 papers
cs.AR2022
LightNorm: Area and Energy-Efficient Batch Normalization Hardware for On-Device DNN Training
Seock-Hwan Noh, Junsang Park, Dahoon Park +3
When training early-stage deep neural networks (DNNs), generating intermediate features via convolution or linear layers occupied most of the execution time. Accordingly, extensive…
cs.LG2022
FlexBlock: A Flexible DNN Training Accelerator with Multi-Mode Block Floating Point Support
Seock-Hwan Noh, Jahyun Koo, Seunghyun Lee +2
Training deep neural networks (DNNs) is a computationally expensive job, which can take weeks or months even with high performance GPUs. As a remedy for this challenge, community h…