J
Jungwook Choi
Researcher at Hanyang University
Publications - 111
Citations - 2622
Jungwook Choi is an academic researcher from Hanyang University. The author has contributed to research in topics: Computer science & Deep learning. The author has an hindex of 17, co-authored 80 publications receiving 1628 citations. Previous affiliations of Jungwook Choi include University of Illinois at Urbana–Champaign & IBM.
Papers
More filters
Posted Content
PACT: Parameterized Clipping Activation for Quantized Neural Networks
Jungwook Choi,Zhuo Wang,Swagath Venkataramani,Pierce I.Jen Chuang,Vijayalakshmi Srinivasan,Kailash Gopalakrishnan +5 more
TL;DR: It is shown, for the first time, that both weights and activations can be quantized to 4-bits of precision while still achieving accuracy comparable to full precision networks across a range of popular models and datasets.
Proceedings Article
Training Deep Neural Networks with 8-bit Floating Point Numbers
TL;DR: In this paper, the authors demonstrate the successful training of deep neural networks using 8-bit floating point numbers while fully maintaining the accuracy on a spectrum of deep learning models and datasets.
Proceedings Article
Hybrid 8-bit Floating Point (HFP8) Training and Inference for Deep Neural Networks
Xiao Sun,Jungwook Choi,Chia Yu Chen,Naigang Wang,Swagath Venkataramani,Vijayalakshmi Srinivasan,Xiaodong Cui,Wei Zhang,Kailash Gopalakrishnan +8 more
TL;DR: This work proposes a hybrid FP8 (HFP8) format and DNN end-to-end distributed training procedure and demonstrates, using HFP8, the successful training of deep learning models across a whole spectrum of applications including Image Classification, Object Detection, Language and Speech without accuracy degradation.
Posted Content
Training Deep Neural Networks with 8-bit Floating Point Numbers
TL;DR: This work demonstrates, for the first time, the successful training of deep neural networks using 8-bit floating point numbers while fully maintaining the accuracy on a spectrum of deep learning models and datasets.
Proceedings Article
AdaComp: Adaptive residual gradient compression for data-parallel distributed training
TL;DR: This paper introduces a novel technique - the Adaptive Residual Gradient Compression (AdaComp) scheme, based on localized selection of gradient residues and automatically tunes the compression rate depending on local activity.