Qi (Cheems) Wang
Associate Professor in Applied Mathematics
BIOGRAPHY
My name is Qi Wang (You can call me Cheems in English). I am currently a Postdoctral researcher at Tsinghua University, working closely with Prof. Xiangyang Ji. I finished the machine learning Ph.D. project at Amsterdam Machine Learning Lab (AMLab), University of Amsterdam. Thanks to my great Ph.D. supervisors, Prof. Max Welling and Dr. Herke van Hoof, for supporting me in finishing this project.
Moreover, my research focuses on the intersection of Meta Learning and Reinforcement Learning. The principal goal of my Ph.D. project is to achieve convincing uncertainty quantification and enable skill transfer across tasks for fast deployment. I have published 3 ICML papers, 3 NeurIPS papers, 1 ICLR paper, X papers under review, and XX papers in progress. My PhD thesis " Functional Representation Learning for Uncertainty Quantification and Fast Skill Transfer " is available in the link, together with the defense video in the link.
More interesting work is ongoing and please follow my updates though Googlescholar (Only selected publications appear in the googlescholar). For any guys who want to achieve scientific collaboration in publishing interesting papers, feel free to contact me😎.
Recent News and Publications
May 17th: The research work "Balanced Confidence Calibration for Graph Neural Networks" was accepted to KDD2024, congratulations to Hao & Min! Predictive confidence matters when deploying graph neural networks in real-world scenarios. This work theoretically analyzes the underconfidence and overconfidence issues in GNN's prediction and develops a balanced calibration approach to address the problem.
May 1st: The research work "Reducing Fine-Tuning Memory Overhead by Approximate and Memory-Sharing Backpropagation" was accepted to ICML2024, congratulations to all my awesome collaborators! Here, a surrogate back-propagation algorithm is developed together with a variant of Layer Normalization. This work significantly reduces memory usage on Lima and ViT and provides theoretical guarantees. Codes will be released soon!
April 2024: For researchers interested in Multi-task Learning, you cannot miss this wonderful work🥳, "GO4Align: Group Optimization for Multi-Task Alignment". Here, we align the learning progress across tasks to promote the exploitation of beneficial task-relatedness and achieve SOTA performance in most existing benchmarks. Codes will be released soon!
October 2023: I joined Tsinghua University and working as a postdoc researcher in automation. Thanks for Prof. Xiangyang Ji's supervision.
September 22nd 2023: Congratulations to Jiayi Shen! Our manuscript "Episodic Multi-Task Learning with Heterogeneous Neural Processes" was accepted to NeurIPS2023 Spotlight (top 3.06% in all submissions), and this work proposes heterogeneous neural processes to exploit task relatedness in multi-task learning. [Paper Link, Code Link]
September 22nd 2023: Our manuscript "A Simple Yet Effective Strategy to Robustify the Meta Learning Paradigm" was accepted to NeurIPS2023, and this work proposes a solution with an improvement guarantee to probabilistically reducing the worst-case risk without sacrificing the average performance. [Paper Link, Slides Link]
August 6th 2023: Our manuscript "Large-scale Generative Simulation Artificial Intelligence: the Next Hotspot" was accepted to the Innovation journal and the flowchart can be found at the top of this section.
July 2023: Many thanks to Max and Herke's nomination and Sara, Yang, Sihang's recommendation, I am honored to be awarded China Computer Federation Multi-agent System Groups (CCF-CMAS) Best Dissertation Prize [中国计算机学会-多智能体系统学组优博论文奖].
January 21st 2023: Our manuscript "Bridge the Inference Gaps of Neural Processes via Expectation Maximization" was accepted to ICLR2023.
October 10th 2022: Very pleased that I received NeurIPS Scholar Award, and thanks to the NeurIPS Foundation.
September 14th 2022: Our paper Learning Expressive Meta-Representations with Mixture of Expert Neural Processes was accepted to NeurIPS2022. Our method applies to few-shot supervised learning and meta reinforcement learning. It is able to handle stochastic processes with mixture components. [Paper Link]
July 2022: The draft of my Ph.D. Thesis will be finished.
July 7th 2022: Delighted to receive ICML2022 Participation Grants.
May 2022: Our paper Model-based Meta Reinforcement Learning using Graph Structured Surrogate Models and Amortized Policy Search was accepted in ICML2022 Spotlight, and the final version will be released soon. It seems the log-likelihood of defending my Ph.D. thesis this year was increased!
Here a GNN based dynamics model is introduced with superior generalization, and the posterior sampling strategy is used in policy learning without additional policy gradients in new environments.
Feel free to access our slides link below as a brief introduction to our proposed GSSM and Amortized Meta Model-based Policy Search. (Note that this is the first trial in amortizing task-specific policies in meta model-based policy search as far as we know. The significance is that the use of non-parametric modeling avoids re-training or gradient adaptations of policies in new environments, which might be promising to address data-efficient fast adaptation problems.)
[The latest version can be found here: Paper Link, Slides Link]
June 1st, 2020 : Our paper entitled "Doubly Stochastic Variational Inference for Neural Processes with Hierarchical Latent Variables" is accepted in ICML2020. (Supplementary materials refer to the Link.)
The slides for our ICML presentation are attached to the Link below. We propose a hierarchical neural process to simultaneously identify tasks and capture local correlations in high-dimensional problems.
Education Background
Before joining AMLab, I obtained a Bachelor's degree in Mathematics at Sichuan University (2011~2015), and after that, I pursued a Master's degree in Management Science at another research institute (2015~2017). During my undergraduate and early graduate periods, I read a lot about statistical learning theory, convex optimization, probabilistic modeling, and risk management in decision-making. Between November 2018 and April 2019, I worked in CSL at UvA. I always feel very grateful to my host, Prof. Peter Sloot, who sincerely supported me at the beginning of my life and research in the Netherlands, especially during my toughest time. At the end of June 2019, I started working at AMLab, under supervison of Prof. Max Welling and Prof. Herke van Hoof.
Research Focus
The source of Uncertainty and the law of Dynamics are of our great concerns in understanding the complexity of the world, and I am fascinated with some novel Bayesian models in learning dynamics and conducting policy optimization in reinforcement learning environments with Statistics and Physics as fundamental techniques.
Currently, I focus more attention on meta learning and reinforcement learning.
Service Updates
Conference Review: ICML/NeurIPS/ICLR
Journal Review: XXX
In August 2021, I feel honored to become a program committee member for the 2021 NeurIPS workshop ECOLOGICAL THEORY OF RL (EcoRL 2021).
Since August 2021, I have been working as a teaching assistant in a Master AI course - Reinforcement Learning at UvA.
Between July 2020 and September 2021, I am assisting our supervisors at AMLab in organizing Weekly Seminar, which is a wonderful platform for research communications.
Since August 2020, I worked as a teaching assistant in a Master AI course - Reinforcement Learning at UvA.
Since August 2019, I worked as a teaching assistant in a Master AI course - Reinforcement Learning at UvA, mainly in charge of practical sessions and QA sessions.
Prepared Study Materials for MSc Students
Based on requests from MSc students in my RL practical sessions, I share my prepared slides of practical sessions with you guys as follows.
Note that these slides are adapted from Richard Sutton's book and other open access online materials.
[session1, session2, session3, session4, session5, session6, session7, session8, session9, session10, session11, session12, to be continued]
Student Supervision
Out-of-Distribution Detection on Time Series Dataset using Bayesian Deep Learning (Collaborate with Fraudio Company) --> Master Student: Berend Jansen --> Finished (2020.12~2021.07)
Contact Information
Office : Lab42 4.22, Science Park 904, Amsterdam
E-mail : q.wang3@uva.nl or hhq123go@gmail.com
Social Media : Twitter @AlbertW24045555
Googlescholar : Q. Wang