I no longer update this website. You will be redirected to my new homepage.

Renjie Liao

I'm a Visiting Faculty Researcher at Google Brain, working with Geoffrey Hinton and David Fleet.
I'm also affiliated with Vector Institute. I will join UBC ECE and CS (associated) as an assistant professor in Jan. 2022.

I'm interested in machine learning and its interplay with vision, self-driving, language, and other areas. In the long run, I am keen to build learning systems that excel at integrating perception and reasoning. Recently, I've been focusing on deep learning for structures, e.g., graphs.

I received my PhD from UofT, advised by Richard Zemel and Raquel Urtasun. During my PhD, I also worked as a research scientist at Uber ATG. I obtained my MPhil from CUHK, advised by Jiaya Jia. I got my BEng from Beihang.

CV

Email: rjliao@google.com rjliao@cs.toronto.edu

"There is nothing as practical as a good theory." - Kurt Lewin
"The test of all knowledge is experiment." - Richard Feynman

News (More)

Our work LookOut: Diverse Multi-Future Prediction and Planning for Self-Driving was accepted (Oral) by ICCV 2021
Congratulations to Alex!
Our work LaneRCNN: Distributed Representations for Graph-Centric Motion Forecasting was accepted by IROS 2021
Congratulations to Wenyuan!
Our work Safety-Oriented Pedestrian Occupancy Forecasting was accepted by IROS 2021
Congratulations to Katie!
I was invited to give a talk "On the Generalization of Graph Neural Networks and their Applications in Probabilistic Inference" at the 2021 Annual Meeting of Canadian Applied and Industrial Math Society (CAIMS)
Our work NP-DRAW: A Non-parametric Structured Latent Variable Model for Image Generation was accepted by UAI 2021
Congratulations to Xiaohui! Code is available!
Our work Nonlinear Equation Solving: A Faster Alternative to Feedforward Computation was accepted by ICML 2021
Congratulations to Yang! Code is available!
Our work Structure-Coherent Deep Feature Learning for Robust Face Alignment was accepted by IEEE TIP 2021
Congratulations to Chunze!
I successfully defended my PhD on Dec. 18 2020 (Dept.) and Mar. 19 2021 (SGS)
Thanks to my committee: Richard Zemel, Raquel Urtasun, Sanja Fidler, David Duvenaud, Murat Erdogdu, and Stefanie Jegelka
My thesis is online!
Our work A PAC-Bayesian Approach to Generalization Bounds for Graph Neural Networks was accepted by ICLR 2021

Research Highlights:

Generalization bounds for GNNs: PAC-Bayes Bounds for GNNs (ICLR 2021)
Deep generative models of graphs: Graph Recurrent Attention Networks (NeurIPS 2019)
Multi-scale spectral graph convolutional networks: LanczosNet (ICLR 2019)
Implicit differentiation: Improving Recurrent Back-Propagation (ICML 2018)
Differentiable bipartite matching: DMM-Net (ICCV 2019)

Last Updated: Jul. 23, 2021