Demonstration venue · read-only. Every page can be browsed; the buttons that would change it are switched off. Create an account to run TaxoReview on your own data. A simulation from public publication records: the people listed were not invited and did not agree to review.
? Tutorial for this page
Demo Data from EDBT 2026 / reviewers in the wild / expert

Jonathan Hseu

dblp:05/3098 · DBLP profile ↗
← Back
2ranked-venue papers
0as first author
0since 2021 · last 2020
—ORCID · none

Domains — the database's venue-derived domains; a paper can count in several

Artificial intelligence and machine learning · 1Systems, architecture and hardware · 1

Expertise — from the expertise taxonomy: the topics of the expert's papers under the CCF categories. A weight counts papers with recency: 1 for a paper about the topic, 0.3 when the topic is its context, halved every five years.

Artificial intelligence
2 papers
Deep learning architectures and training · 55% Efficient and distributed learning · 25% Optimization for machine learning · 20%

Topics — the 4 heaviest of 4, each with the papers that count most for it

TopicWeightPapersLastEvidence papers
Machine learning › Deep learning architectures and training › training optimization
large-batch training		 0.822020
Large Batch Optimization for Deep Learning: Training BERT in 76 minutes · ICLR 2020
Large-batch training for LSTM and beyond · SC 2019
Machine learning › Efficient and distributed learning
distributed training		 0.522020
Large Batch Optimization for Deep Learning: Training BERT in 76 minutes · ICLR 2020
Large-batch training for LSTM and beyond · SC 2019
Machine learning › Optimization for machine learning › large-scale optimization
large batch optimization		 0.412020
Large Batch Optimization for Deep Learning: Training BERT in 76 minutes · ICLR 2020
Machine learning › Deep learning architectures and training
recurrent neural network		 0.412019
Large-batch training for LSTM and beyond · SC 2019

Methods — techniques the papers use, named apart from their topics

layer-wise adaptive learning rate · 0.4sqrt scaling · 0.4dynamic adaptive-tuning engine · 0.4
YearPublicationVenuePosition
2020 Large Batch Optimization for Deep Learning: Training BERT in 76 minutes
Yang You 0001, Sashank J. Reddi, Jonathan Hseu, Sanjiv Kumar, Srinadh Bhojanapalli, Xiaodan Song, James Demmel, Kurt Keutzer, Cho-Jui Hsieh
 ICLR4
2019 Large-batch training for LSTM and beyond
abstract
Large-batch training approaches have enabled researchers to utilize distributed processing and greatly accelerate deep neural networks training. However, there are three problems in current large-batch research: (1) Although RNN approaches like LSTM have been widely used in many applications, current large-batch research is principally focused on CNNs. (2) Even for CNNs, there is no automated technique for extending the batch size beyond 8K. (3) To keep the variance in the gradient expectation constant, theory suggests that a Sqrt Scaling scheme should be used in large-batch training. Unfortunately, there are not many successful applications. In this paper, we propose Dynamic Adaptive-Tuning Engine (DATE) for better large-batch training. DATE achieves a 5.3x average speedup over the baselines for four LSTM-based applications on the same hardware. We finish the ImageNet training with ResNet-50 in two minutes on 1024 v3 TPUs (76.7% top-1 accuracy), which is the fastest version as of June of 2019.
Yang You 0001, Jonathan Hseu, Chris Ying, James Demmel, Kurt Keutzer, Cho-Jui Hsieh
 SC2