Anqi
Wu

General Information

Email:
awu36@gatech.edu
Phone:
3238681604
Location - Building:
Coda
Location - Room:
S1311
Roles:
Professor (any rank)
Primary Unit:
School of Computational Science and Engineering

Details

Degrees with subject and Postdoc Experience:
Degree Type
postdoctoral scholar
Subject
Computational Neuroscience
Year
2019-2021
Institution
Columbia University
Location
New York City
Degree Type
Ph.D.
Subject
Computational Neuroscience
Year
2019
Institution
Princeton University
Location
Princeton
Degree Type
M.S.
Subject
Computer Science
Year
2014
Institution
University of Southern California
Location
Los Angeles
Degree Type
B.S.
Subject
Electronics and Information Engineering
Year
2011
Institution
Harbin Institute of Technology
Location
Harbin, China
Statement of Research Interests:

The lab focuses on developing probabilistic modeling approaches and scalable and efficient inference algorithms, with applications to neural and behavior analyses, as well as many real-world problems.

More specifically, the modeling topics involve: deep generative models, variational autoencoder, (deep) Gaussian process, Bayesian (convolutional) neural net, Bayesian nonparametric, Bayesian optimization and active learning, computer vision, hierarchical spatial and temporal models, latent dynamic models, (inverse) reinforcement learning, etc.

The applications cover but are not limited to: neural latent discovery, 3D full-body kinematic model estimation, identifying behavior syllables, studying intrinsic motives and reward representations of animal and human behaviors, fMRI decoding, neural sensory encoding, optimal experimental design, etc.

Statement of Teaching Interests:

I take an iterative and reflective approach to teaching and continually refine my instructional methods to improve clarity, effectiveness, and student engagement. I have taught CSE 6740 and CSE 8803: Machine Learning for Neural and Behavioral Data Analysis, where I emphasize helping students understand how different models relate to one another rather than treating methods as isolated techniques. My teaching goal is to support students in building a coherent and systematic understanding of data analysis and machine learning, enabling them to confidently apply these tools in new settings. I encourage active learning through in-class questions and discussion and regularly update course materials to incorporate emerging methods and newly developed techniques, an approach that has been positively reflected in student feedback and teaching recognition.

Selection of recent research, scholarly, and creative activities:

Selected Awards & Recognition

2025 Kavli Fellow by National Academy of Sciences

2023 Arab-American Frontiers of Science, Engineering, and Medicine initiated by the National Academies

2023 Alfred P. Sloan Fellow in Neuroscience

2022 DARPA Riser

Selected Publications

Uncovering Semantic Selectivity of Latent Groups in Higher Visual Cortex with Mutual Information-Guided Diffusion
International Conference on Learning Representations (ICLR) (acceptance rate: 28%)
Yule Wang, Joseph Yu, Chengrui Li, Weihan Li, Anqi Wu — 2026

Learn to Teach: Sample-Efficient Privileged Learning for Humanoid Locomotion over Real-World Uneven Terrain
IEEE Robotics and Automation Letters (RA-L)
Feiyang Wu, Xavier Nal, Jaehwi Jang, Wei Zhu, Zhaoyuan Gu, Anqi Wu, Ye Zhao — 2025

Inverse Reinforcement Learning with Switching Rewards and History Dependency for Characterizing Animal Behaviors
International Conference on Machine Learning (ICML) (acceptance rate: 26.9%)
Jingyang Ke, Feiyang Wu, Jiyi Wang, Jeffrey Markowitz, Anqi Wu — 2025

Learning Time-Varying Multi-Region Communications via Scalable Markovian Gaussian Processes
International Conference on Machine Learning (ICML) (Oral, 1%)
Weihan Li, Chengrui Li, Yule Wang, Anqi Wu — 2024

Exploring Behavior-Relevant and Disentangled Neural Dynamics with Generative Diffusion Models
Conference on Neural Information Processing Systems (NeurIPS) (acceptance rate: 25.8%)
Yule Wang, Chengrui Li, Weihan Li, Anqi Wu — 2024

Inverse Kernel Decomposition
Transactions on Machine Learning Research (TMLR)
Chengrui Li, Anqi Wu — 2024

A Differentiable POGLM with Forward–Backward Message Passing
International Conference on Machine Learning (ICML) (acceptance rate: 27.5%)
Chengrui Li, Weihan Li, Yule Wang, Anqi Wu — 2024

Multi-Region Markovian Gaussian Process: An Efficient Method to Discover Directional Communications Across Multiple Brain Regions
International Conference on Machine Learning (ICML) (acceptance rate: 27.5%)
Weihan Li, Chengrui Li, Yule Wang, Anqi Wu — 2024

Infer and Adapt: Bipedal Locomotion Reward Learning from Demonstrations via Inverse Reinforcement Learning
International Conference on Robotics and Automation (ICRA)
Feiyang Wu, Zhaoyuan Gu, Hanran Wu, Anqi Wu, Ye Zhao (co-senior authorship) — 2024

Forward χ² Divergence–Based Variational Importance Sampling
International Conference on Learning Representations (ICLR) (Spotlight, 5%)
Chengrui Li, Yule Wang, Weihan Li, Anqi Wu — 2024

One-Hot Generalized Linear Model for Switching Brain State Discovery
International Conference on Learning Representations (ICLR) (acceptance rate: 31%)
Chengrui Li, Soon Ho Kim, Chris Rodgers, Hannah Choi, Anqi Wu — 2024