Zhongwen Xu

Principal Scientist
Sea AI Lab

[Google Scholar] [DBLP] [Linkedin]

Email:

Bio. I am a Principal Scientist at Sea AI Lab, focusing on Deep Reinforcement Learning. I was a senior research scientist at DeepMind before joining Sea. I pursued my Ph.D. at University of Technology Sydney (UTS), advised by Prof. Yi Yang. I received my Bachelor's degree from Zhejiang University in 2013, under the supervision of Prof. Yueting Zhuang and Prof. Fei Wu.

I spent two fabulous internships in Alphabet/Google's AI groups. In Spring 2017, I was a research intern at DeepMind, working with Dr. Hado van Hasselt and Prof. David Silver. And in Spring 2016, I was a research intern at Google Brain, working with Dr. Pierre Sermanet and Dr. George Toderici.

I visited Carnegie Mellon University in 2012 and 2014, hosted by Dr. Alexander G. Hauptmann. In 2014, I visited National University of Singapore, working with Prof. Shuicheng Yan.

More details can be found in my CV.

We're hiring, please reach out if you're interested in working with us.

PUBLICATIONS

Emphatic Algorithms for Deep Reinforcement Learning
Ray Jiang, Tom Zahavy, Zhongwen Xu, Adam White, Matteo Hessel, Charles Blundell, Hado van Hasselt 1
ICML 2021

Meta-Gradient Reinforcement Learning with an Objective Discovered Online
Zhongwen Xu, Hado van Hasselt, Matteo Hessel, Junhyuk Oh, Satinder Singh, David Silver
NeurIPS 2020

Balancing Constraints and Rewards with Meta-Gradient D4PG
Dan A Calian, Daniel J Mankowitz, Tom Zahavy, Zhongwen Xu, Junhyuk Oh, Nir Levine, Timothy Mann
ICLR 2021

Discovering Reinforcement Learning Algorithms
Junhyuk Oh, Matteo Hessel, Wojciech M. Czarnecki, Zhongwen Xu, Hado van Hasselt, Satinder Singh, David Silver
NeurIPS 2020

A Self-Tuning Actor-Critic Algorithm
Tom Zahavy, Zhongwen Xu, Vivek Veeriah, Matteo Hessel, Junhyuk Oh, Hado van Hasselt, David Silver, Satinder Singh
NeurIPS 2020

What Can Learned Intrinsic Rewards Capture?
Zeyu Zheng, Junhyuk Oh, Matteo Hessel, Zhongwen Xu, Manuel Kroiss, Hado van Hasselt, David Silver, Satinder Singh
ICML 2020

Discovery of Useful Questions as Auxiliary Tasks
Vivek Veeriah, Matteo Hessel, Zhongwen Xu, Richard Lewis, Janarthanan Rajendran, Junhyuk Oh, Hado van Hasselt, David Silver, Satinder Singh
NeurIPS 2019

Meta-Gradient Reinforcement Learning
Zhongwen Xu, Hado van Hasselt, and David Silver
NeurIPS 2018 [Poster]

Watching a Small Portion could be as Good as Watching All: Towards Efficient Video Classification
Hehe Fan, Zhongwen Xu, Linchao Zhu, Chenggang Yan, Jianjun Ge, and Yi Yang
IJCAI 2018

Natural Value Approximators: Learning When to Trust Past Estimates.
Zhongwen Xu, Joseph Modayil, Hado van Hasselt, Andre Barreto, David Silver and Tom Schaul
NIPS 2017 (Spotlight)

Few-Shot Object Recognition from Machine-Labeled Web Images
Zhongwen Xu*, Linchao Zhu* and Yi Yang
CVPR 2017 (Spotlight) [Code] [Spotlight Video]

Bidirectional Multirate Reconstruction for Temporal Modeling in Videos
Linchao Zhu, Zhongwen Xu and Yi Yang
CVPR 2017 (Spotlight) [Code]

Uncovering Temporal Context for Video Question Answering
Linchao Zhu, Zhongwen Xu, Yi Yang and Alexander G. Hauptmann
IJCV 2017 [arXiv] [Dataset]

An End-to-End Approach to Natural Language Object Retrieval via Context-Aware Deep Reinforcement Learning
Fan Wu, Zhongwen Xu and Yi Yang
arXiv preprint 1703.07579 [Code]

Hierarchical Recurrent Neural Encoder for Video Representation with Application to Captioning
Pingbo Pan, Zhongwen Xu, Yi Yang, Fei Wu and Yueting Zhuang
CVPR 2016 [PDF]

Robust Semi-supervised Learning through Label Aggregation
Yan Yan, Zhongwen Xu, Ivor W. Tsang, Guodong Long and Yi Yang
AAAI 2016 [PDF]

A Discriminative CNN Video Representation for Event Detection
Zhongwen Xu, Yi Yang and Alexander G. Hauptmann
CVPR 2015 [PDF]

Content-Based Video Search over 1 Million Videos with 1 Core in 1 Second
Shoou-I Yu, Lu Jiang, Zhongwen Xu, Yi Yang and Alexander G. Hauptmann
ICMR 2015

Event Detection Using Multi-level Relevance Labels and Multiple Features
Zhongwen Xu, Ivor W. Tsang, Yi Yang, Zhigang Ma and Alexander G. Hauptmann
CVPR 2014 [PDF]

Feature Weighting via Optimal Thresholding for Video Analysis
Zhongwen Xu, Yi Yang, Ivor W. Tsang, Nicu Sebe and Alexander G. Hauptmann
ICCV 2013 [PDF]

How Related Exemplars Help Complex Event Detection in Web Videos?
Yi Yang, Zhigang Ma, Zhongwen Xu, Shuicheng Yan and Alexander G. Hauptmann
ICCV 2013 [PDF]

Complex Event Detection via Multi-source Video Attributes
Zhigang Ma, Yi Yang, Zhongwen Xu, Shuicheng Yan, Nicu Sebe and Alexander G. Hauptmann
CVPR 2013 [PDF]

We Are Not Equally Negative: Fine-grained Labeling for Multimedia Event Detection
Zhigang Ma, Yi Yang, Zhongwen Xu, Nicu Sebe and Alexander G. Hauptmann
ACM Multimedia 2013 [PDF]


PATENTS

Meta-gradient updates for training return functions for reinforcement learning systems
Zhongwen Xu, Hado van Hasselt, David Silver
US Patent App. 16/417,536

Training action selection neural networks using a differentiable credit function
Zhongwen Xu, Hado van Hasselt, Joseph Modayil, Andre Barreto, David Silver
US Patent App. 16/615,042

COMPETITIONS

UTS-CMU at THUMOS 2015
Zhongwen Xu, Linchao Zhu, Yi Yang and Alexander G. Hauptmann
THUMOS challenge 2015
[Ranked 1st place] [PDF] [THUMOS Challenge]

Cross-media Relevance Mining for Evaluating Text-based Image Search Engine
Zhongwen Xu, Yi Yang, Ashraf A. Kassim and Shuicheng Yan
ICME MSR-Bing Grand Challenge Workshop 2014
[Ranked 1st place] [PDF] [MSR-Bing Image Retrieval Challenge (IRC)]

Informedia@TRECVID 2014 MED and MER
Shoou-I Yu, Lu Jiang, Zhongwen Xu, et al. and Alexander G. Hauptmann
TRECVID Workshop 2014
[Ranked 1st place in MED] [PDF] [TRECVID Multimedia Event Detection 2014]

ILSVRC (ImageNet) 2014 Classification with Provided Data Only
Zhongwen Xu and Yi Yang
[Ranking: Google, VGG, MSRA, Howard, DeepVision, NUS, TTIC, XYZ (ours)] [link]

Informedia E-Lamp @ TRECVID 2012 Multimedia Event Detection and Recounting (MED and MER)
Shoou-I Yu, Zhongwen Xu, Duo Ding, et al. and Alexander G. Hauptmann
TRECVID Workshop 2012
[Ranked 1st place in Pre-specific events, 2nd place in Ad-hoc events] [PDF]
[TRECVID Multimedia Event Detection 2012]


Theme from Karen Simonyan.