H
Huyen Nguyen
Researcher at Nvidia
Publications - 10
Citations - 545
Huyen Nguyen is an academic researcher from Nvidia. The author has contributed to research in topics: Language model & Acoustic model. The author has an hindex of 6, co-authored 10 publications receiving 350 citations.
Papers
More filters
Proceedings ArticleDOI
Jasper: An End-to-End Convolutional Neural Acoustic Model.
Jason Li,Vitaly Lavrukhin,Boris Ginsburg,Ryan Leary,Oleksii Kuchaiev,Jonathan Cohen,Huyen Nguyen,Ravi Teja Gadde +7 more
TL;DR: This paper reports state-of-the-art results on LibriSpeech among end-to-end speech recognition models without any external training data and introduces a new layer-wise optimizer called NovoGrad to improve training.
Posted Content
NeMo: a toolkit for building AI applications using Neural Modules.
Oleksii Kuchaiev,Jason Li,Huyen Nguyen,Oleksii Hrinchuk,Ryan Leary,Boris Ginsburg,Samuel Kriman,Stanislav Beliaev,Vitaly Lavrukhin,Jack Cook,Patrice Castonguay,Mariya Popova,Jocelyn Huang,Jonathan Cohen +13 more
TL;DR: NeMo (Neural Modules) is a Python framework-agnostic toolkit for creating AI applications through re-usability, abstraction, and composition that provides built-in support for distributed training and mixed precision on latest NVIDIA GPUs.
Posted Content
Jasper: An End-to-End Convolutional Neural Acoustic Model
Jason Li,Vitaly Lavrukhin,Boris Ginsburg,Ryan Leary,Oleksii Kuchaiev,Jonathan Cohen,Huyen Nguyen,Ravi Teja Gadde +7 more
TL;DR: Jasper as mentioned in this paper uses only 1D convolutions, batch normalization, ReLU, dropout, and residual connections to improve training, and further introduces a new layer-wise optimizer called NovoGrad.
Posted Content
Stochastic Gradient Methods with Layer-wise Adaptive Moments for Training of Deep Networks
Boris Ginsburg,Patrice Castonguay,Oleksii Hrinchuk,Oleksii Kuchaiev,Vitaly Lavrukhin,Ryan Leary,Jason Li,Huyen Nguyen,Jonathan Cohen +8 more
TL;DR: NovoGrad, an adaptive stochastic gradient descent method with layer-wise gradient normalization and decoupled weight decay, performs on par or better than well tuned SGD with momentum and Adam or AdamW in experiments on neural networks.
Posted Content
Mixed-Precision Training for NLP and Speech Recognition with OpenSeq2Seq
Oleksii Kuchaiev,Boris Ginsburg,Igor Gitman,Vitaly Lavrukhin,Jason Li,Huyen Nguyen,Carl Case,Paulius Micikevicius +7 more
TL;DR: Benchmarks on machine translation and speech recognition tasks show that models built using OpenSeq2Seq give state-of-the-art performance at 1.5-3x less training time.