Zhouyuan Huo
I am a senior research scientist at Google. I received Ph.D. degree in Electrical & Computer Engineering at University of Pittsburgh under the supervision of Prof. Heng Huang.
Before that, I received B.Eng degree at Zhejiang University in 2014. My CV can be found here.
My recent research interest lies at the intersection of machine learning and distributed optimization, especially federated learning and self-supervised learning.
Email: huozhouyuan[at]gmail[dot]com
Q
Invited to serve on an NSF Panel.
One paper was accepted by NeurIPS 2021.
Invited to serve as a session chair for KDD 2021 (Deep Learning Applications).
One paper was accepted by Interspeech.
Invited to speak at MIT Virtual Workshop on Split Learning for Distributed Machine Learning (SLDML’21).
One paper was accepted by TNNLS.
One paper was accepted by TPAMI.
Two papers were accepted by AAAI 2021.
Serve as a Senior Program Committee (SPC) for IJCAI 2021.
One paper was accepted by JMLR.
Serve as a session Chair for KDD 2020 (Parallel and Distributed Learning and System).
One paper was accepted by CVPR 2020.
One paper was accepted by NeurIPS 2019.
One paper was accepted by EMNLP 2019.
Large Batch Optimization for Deep Learning Using New Complete Layer-Wise Adaptive Rate Scaling
AAAI 2021
[Paper]
[Code]
Step-Ahead Error Feedback for Distributed Training with Compressed Gradient
AAAI 2021
[Paper]
A Unified q-Memorization Framework for Asynchronous Stochastic Optimization
JMLR
[Paper]
On the Acceleration of Deep Learning Model Parallelism with Staleness
CVPR 2020
[Paper]
On Accelerating Training of Transformer-Based Language Models
NeurIPS 2019
[Paper]
An End-to-End Generative Architecture for Paraphrase Generation
EMNLP 2019
[Paper]
Visual-GPS: Ego-Downward and Ambient Video Based Person Location Association
CVPR Workshop 2019
[Paper]
Scalable and Efficient Pairwise Learning to Achieve Statistical Accuracy
AAAI 2019
[Paper]
Faster Gradient-Free Proximal Stochastic Methods for Nonconvex Nonsmooth Optimization
AAAI 2019
[Paper]
Training Neural Networks Using Features Replay
NeurIPS 2018 (Spotlight)
[Paper]
[Code]
Asynchronous Dual Free Stochastic Dual Coordinate Ascent for Distributed Data Mining
ICDM 2018
[Paper]
Decoupled Parallel Backpropagation with Convergence Guarantee
ICML 2018 (Oral)
[Paper]
[Code]
Faster Derivative-Free Stochastic Algorithm for Shared Memory Machines
ICML 2018 (Oral)
[Paper]
Asynchronous Doubly Stochastic Group Regularized Learning
AISTATS, 2018.
[Paper]
[Code]
Accelerated Method for Stochastic Composition Optimization with Nonsmooth Regularization
AAAI, 2018.
[Paper]
[Code]
Inexact Proximal Gradient Methods for Non-Convex and Non-Smooth Optimization
AAAI, 2018.
[Paper]
Asynchronous Doubly Stochastic Sparse Kernel Learning
AAAI, 2018.
[Paper]
New Multi-Task Learning Model to Study the Genotype-Phenotype Associations
PSB, 2018.
[Paper]
Joint Capped Norms Minimization for Robust Matrix Recovery
IJCAI, 2017.
[Paper]
Maximizing Multi-Class Margins for Supervised and Semi-Supervised Support Vector Machine
IJCAI, 2017.
[Paper]
Asynchronous Mini-Batch Gradient Descent with Variance Reduction for Non-Convex Optimization
AAAI, 2017.
[Paper]
[Code]
Video Recovery via New Sectional Trace Norm with Capped Norm Constrained Variation and Consistency
AAAI, 2017.
[Paper]
New Probabilistic Multi-Graph Decomposition Model to Identify Consistent Human Brain Network Models
ICDM, 2016.
[Paper]
Robust and Effective Metric Learning Using Capped Trace Norm: Metric Learning via Capped Trace Norm
KDD, 2016.
[Paper]
New Multi-task Learning Model to Predict Alzheimer’s Disease Cognitive Assessment
MICCAI, 2016.
[Paper]
Optimal Discrete Matrix Completion
AAAI, 2016.
[Paper]
[Code]
Senior Program Committee:
IJCAI (2021)
Program Committee:
NSF panel, NeurIPS (2016, 18-20), ICLR (2019-20), ICML (2019-21), IJCAI (2018-20), AAAI (2018-21), CVPR (2019-21), ICCV (2019), ECCV (2020), KDD (2020-21), ACL (2021)
Journals Reviewer:
Journal of Machine Learning Research (JMLR), IEEE Transactions on Knowledge and Data Engineering (TKDE), IEEE Transactions on Neural Networks and Learning Systems (TNNLS), IEEE Transactions on Signal Processing, Information Sciences