Yujia Li


Profile Picture


Department of Computer Science
University of Toronto
UofT Email
I'm a Research Scientist at DeepMind. I have obtained my Ph.D. from University of Toronto. I was a member of the machine learning group and my advisor was Prof. Richard Zemel. Before coming to Toronto, I obtained my bachelor's degree from Tsinghua University, Beijing, China. I was an intern Research and Development engineer at Baidu, Inc. in the summer of 2011, a Research Intern with the Speech Group at Microsoft Research Redmond in the summer of 2014, and a Research Intern with the Machine Learning and Perception group at Microsoft Research Cambridge in the summer of 2015.


News

  • 2 new papers accepted to appear in NIPS 2017, one Oral and one Spotlight.
  • 3 new papers on arXiv.
  • I joined DeepMind as a Research Scientist!
  • I defended my thesis and finally graduated!
  • One paper accepted to appear in NIPS 2016.
  • Code for our ICLR 2016 paper "Gated Graph Sequence Neural Networks" is available now.
  • Two papers accepted to appear in ICLR 2016.
  • One paper accepted to appear in ICML 2015.
  • One paper accepted to appear in ICASSP 2015.


Education

2013.2-2017.2      Ph.D. Student
Department of Computer Science, University of Toronto, Toronto, ON, Canada
 
2011.9-2013.1      Master of Science
Department of Computer Science, University of Toronto, Toronto, ON, Canada
 
2007.8-2011.7 Bachelor of Engineering
School of Software, Tsinghua University, Beijing, China


Publications

Preprints

Learning Model-Based Planning from Scratch
Razvan Pascanu*, Yujia Li*, Oriol Vinyals, Nicolas Heess, Lars Buesing, Sébastien Racanière, David Reichert, Théophane Weber, Daan Wierstra, Peter Battaglia (* denotes equal contribution)
arXiv:1707.06170
[arXiv]

Conference Papers

Dualing GANs
Yujia Li, Alexander Schwing, Kuan-Chieh Wang and Richard Zemel
Neural Information Processing Systems (NIPS), 2017
Spotlight Presentation
[arXiv]

Imagination-Augmented Agents for Deep Reinforcement Learning
Théophane Weber*, Sébastien Racanière*, David P. Reichert*, Lars Buesing, Arthur Guez, Danilo Jimenez Rezende, Adria Puigdomènech Badia, Oriol Vinyals, Nicolas Heess, Yujia Li, Razvan Pascanu, Peter Battaglia, David Silver, Daan Wierstra (* denotes equal contribution)
Neural Information Processing Systems (NIPS), 2017
Oral Presentation
[arXiv]

Understanding the Effective Receptive Field in Deep Convolutional Neural Networks
Wenjie Luo*, Yujia Li*, Raquel Urtasun, and Richard Zemel (* denotes equal contribution)
Neural Information Processing Systems (NIPS), 2016
[paper] [poster]

Gated Graph Sequence Neural Networks
Yujia Li, Daniel Tarlow, Marc Brockschmidt and Richard Zemel
International Conference on Learning Representations (ICLR), 2016
[arXiv] [slides] [poster] [code] [talk]

The Variational Fair Auto Encoder
Christos Louizos, Kevin Swersky, Yujia Li, Max Welling, Richard Zemel
International Conference on Learning Representations (ICLR), 2016
Oral Presentation
[arXiv]

Generative Moment Matching Networks
Yujia Li, Kevin Swersky and Richard Zemel
International Conference on Machine Learning (ICML), 2015
[paper] [arXiv preprint] [code] [project page]

Feedback-Based Handwriting Recognition from Inertial Sensor Data for Wearable Devices
Yujia Li, Kaisheng Yao and Geoffrey Zweig
The 40th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2015
[paper]

High Order Regularization for Semi-Supervised Learning of Structured Output Problems
Yujia Li and Richard Zemel
International Conference on Machine Learning (ICML), 2014
[paper + supplementary material] [slides] [poster] [data] [code] [video]

Exploring Compositional High Order Pattern Potentials for Structured Output Learning
Yujia Li, Daniel Tarlow and Richard Zemel
The 26th IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2013
Oral Presentation
[paper] [supplementary material] [slides] [poster] [data] [code] [video]

Celebrity Recommendation with Collaborative Social Topic Regression
Xuetao Ding, Xiaoming Jin, Yujia Li and Lianghao Li
Proceedings of the 23rd International Joint Conference on Artificial Intelligence (IJCAI), 2013
[paper]


Workshop Papers

Learning Unbiased Features
Yujia Li, Kevin Swersky and Richard Zemel
NIPS workshop on Transfer and Multi-Task Learning, 2014
[paper] [slides] [poster]

Mean Field Networks
Yujia Li and Richard Zemel
ICML workshop on Learning Tractable Probabilistic Models, 2014
[paper] [slides] [poster]


Thesis

Building More Expressive Structured Models
Ph.D. Thesis
[paper]

Exploring Compositional High Order Pattern Potentials for Structured Output Learning
M.Sc. Thesis
[paper]



Talks

Generative Moment Matching Networks (2015) at Toronto Machine Learning Group Seminar. [pdf]

A Tutorial on Dual Decomposition (2014) at Toronto Machine Learning Group Meeting. [pdf]



Teaching Experience

Courses TA'ed at University of Toronto Courses TA'ed at Tsinghua University
  • Spring 2011, Advanced Data Structures
  • Spring 2011, Machine Learning and Knowledge Discovery


Honors and Awards

  • 2016, ICLR travel award.
  • 2015, Microsoft Ph.D. fellowship (US and Canada) finalist.
  • 2015, ICML travel grant.
  • 2013, CVPR travel grant.
  • 2013, University of Toronto School of Graduate Studies conference grant.
  • 2008-2010, University-wide comprehensive merit scholarship, three times - Including Kai-Feng Scholarship, which is the highest amount among all scholarships and awarded to only 30 undergraduate students in Tsinghua University every year.
  • 2008, 2nd Prize - Chinese National College Physics Contest
  • 2006, 1st Prize - Chinese Physics Olympiad (CPhO) in Provinces
  • 2005, 2nd Prize - Chinese National Olympiad in Informatics in Provinces (NOIP)

Other Stuff

My name in Chinese: My Name

A list of my old projects can be found here.

I swim and run regularly every week. I like playing badminton. I'm a fan of NBA. Following are some activities that I have participated in.

  • 2017, Finished 10km - London Winter Run
  • 2016, Finished Half Marathon - Goodlife fitness Toronto Marathon
  • 2010, Finished Half Marathon - 30th Beijing International Marathon
  • 2010, Volunteer - Data Statistics group, China Open (Tennis)
  • 2009, Visiting Student - Tohoku University, Sendai, Japan.
  • 2009, Vice President - Student Association of Science and Technology, School of Software, Tsinghua University
    • Co-organized the first regional software design contest, which attracts around 70 teams from 7 main universities in Beijing to participate.
  • 2008-2009, Section Chief - Student Union, School of Software, Tsinghua University
  • 2008, Staff, Out Standing Staff Laureate - Media Village, Beijing Olympic Games
  • 2008, Finished 10km - 28th Beijing International Marathon