 |
Assistant Professor of Computer Science Email: hongyang90@gmail.com, or Address: 177 Huntington Ave, Room 2211 (note). |
Hi! I am an assistant professor of computer science at Northeastern University since Fall 2020. I am interested in algorithmic and data-centric aspects of machine learning and network data analysis.
On the machine learning side, I'm interested in methodological and theoretical questions related to ML algorithms and generalization, including:
Generalization in overparametrized models: How can we reason about the generalization error of a deep neural network model? What if an ML algorithm is applied to out-of-distribution data?
Learning with weak supervision: Multitask learning, data augmentation, and contrastive learning for dealing with limited labeled data.
On the network side, I'm interested in the design of algorithms motivated by applications in social networks and epidemiology.
Learning on graphs: How shall we reason about the generalization and robustness properties of graph neural networks?
I came to Northeastern after a brief postdoc study at UPenn. I received my Ph.D. from Stanford, advised by Ashish Goel and Greg Valiant. I also collaborated with Tengyu Ma and Chris Ré's groups at Stanford. I received a B.Eng. from Shanghai Jiao Tong University (the ACM class).
May 2022 (paper): Excited about two recent works that will be presented at ICML’22 (thanks to the anonymous referees for feedback!):
One paper with Michael Zhang et al. about a contrastive learning approach for improving worst-group performance.
Another paper with my students about Hessian based generalization for fine-tuned models.
May 2022 (talk): gave a talk about understanding and improving generalization in multitask and transfer learning at the One World ML Seminar.
Mar 2022 (service): PC member of KDD’22, reviewer of ICML’22.
Mar 2022 (talk): presenting our recent work on fine-tuning at Northeastern's CS Theory Lunch Seminar.
Feb 2022 (preprint): a new paper that proposes generalization measures and robust algorithms for fine-tuning.
Jan 2022 (teaching): advanced machine learning in the spring semester.
A list of old updates
Research
Manuscripts
Analysis of Information Transfer from Heterogeneous Sources via Precise High-dimensional Asymptotics
Fan Yang, H.R. Zhang, Sen Wu, Weijie Su, and Christopher Ré
Working paper, 2021
Publications
Optimal Intervention on Weighted Networks via Edge Centrality
Dongyue Li, Tina Eliassi-Rad, and H. R. Zhang
KDD epiDAMIK Workshop 2022
Task Modeling: A Multitask Approach for Improving Robustness to Group Shifts
Dongyue Li, Huy L. Nguyen, and H. R. Zhang
ICML Principles of Distribution Shift Workshop 2022
Robust Fine-Tuning of Deep Neural Networks with Hessian-based Generalization Guarantees
Haotian Ju, Dongyue Li, and H. R. Zhang
Interntional Conference on Machine Learning (ICML) 2022
[ICML Updatable ML Workshop 2022]
Correct-N-Contrast: A Contrastive Approach for Improving Robustness to Spurious Correlations
Michael Zhang, Nimit Sohoni, H. R. Zhang, Chelsea Finn, and Christopher Ré
International Conference on Machine Learning (ICML) 2022. Long presentation
[NeurIPS DistShift Workshop 2021] [ICML UDL Workshop 2021]
Incentive Ratio: A Game Theoretical Analysis of Market Equilibria
Ning Chen, Xiaotie Deng, Bo Tang, H. R. Zhang, and Jie Zhang
Information and Computation, 2022
Improved Regularization and Robustness for Fine-tuning in Neural Networks
Dongyue Li and H. R. Zhang
Neural Information Processing Systems (NeurIPS) 2021
[code]
Observational Supervision for Medical Image Classification using Gaze Data
Khaled Saab, Sarah Hooper, Nimit Sohoni, Jupinder Parmar, Brian Pogatchnik, Sen Wu, Jared Dunnmon, H. R. Zhang, Daniel Rubin, and Christopher Ré
Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2021
On the Generalization Effects of Linear Transformations in Data Augmentation
Sen Wu*, H. R. Zhang*, Gregory Valiant, and Christopher Ré
International Conference on Machine Learning (ICML) 2020
[blog] [code]
Learning Over-Parametrized Two-Layer ReLU Neural Networks beyond NTK
Yuanzhi Li, Tengyu Ma, and H. R. Zhang*
Conference on Learning Theory (COLT) 2020
Understanding and Improving Information Transfer in Multi-Task Learning
Sen Wu*, H. R. Zhang*, and Christopher Ré
International Conference on Learning Representations (ICLR) 2020
[blog] [video]
Pruning based Distance Sketches with Provable Guarantees on Random Graphs
H. R. Zhang, Huacheng Yu, and Ashish Goel
The Web Conference (WWW) 2019. Oral presentation
Recovery Guarantees for Quadratic Tensors with Limited Observations
H. R. Zhang, Vatsal Sharan, Moses Charikar, and Yingyu Liang
International Conference on AI and Statistics (AISTATS) 2019
Algorithmic Regularization in Over-parameterized Matrix Sensing and Neural Networks with Quadratic Activations
Yuanzhi Li, Tengyu Ma, and H. R. Zhang*
Conference on Learning Theory (COLT) 2018. Best Paper Award
Approximate Personalized PageRank on Dynamic Graphs
H. R. Zhang, Peter Lofgren, and Ashish Goel
KDD 2016. Oral presentation
[code]
Incentives for Strategic Behavior in Fisher Market Games
Ning Chen, Xiaotie Deng, Bo Tang, and H. R. Zhang*
AAAI 2016
A Note on Modeling Retweet Cascades on Twitter
Ashish Goel, Kamesh Munagala, Aneesh Sharma, and H. R. Zhang*
Workshop on Algorithms and Models for the Web Graph (WAW) 2015
Connectivity in Random Forests and Credit Networks
Ashish Goel, Sanjeev Khanna, Sharath Raghvendra, and H. R. Zhang*
Symposium on Discrete Algorithms (SODA) 2015
Computing the Nucleolus of Matching, Cover and Clique Games
Ning Chen, Pinyan Lu, and H. R. Zhang*
AAAI 2012. Oral presentation
Incentive Ratios of Fisher Markets
Ning Chen, Xiaotie Deng, H. R. Zhang*, and Jie Zhang
International Colloquium on Automata, Languages, and Programming (ICALP) 2012
On Strategy-proof Allocation without Payments or Priors
Li Han, Chunzhi Su, Linpeng Tang, and H. R. Zhang*
Conference on Web and Internet Economics (WINE) 2011
Remark: Some of my papers, marked by an asterisk, use an alphabetical ordering of author names. If you find my work interesting, you might also be interested in checking out the following blog posts.
When Multi-Task Learning Works – And When It Doesn’t
based on joint work with Sen Wu and Chris Ré.Automating the Art of Data Augmentation – Part III Theory
edited with Sharon Y. Li and Chris Ré.
Students
Ph.D. students
Dongyue Li (2021)
Haoyu He (incoming 2022)
Master's students
Haotian Ju (2021)
Shreya Singh (2022)
Alumni
Virender Singh (MS 2021; First employment: Salesforce)
Prospective students: I'm always looking for students. If you are already at Northeastern, please feel free to get in touch. We have projects that could use your help: Check out our recent projects at GitHub. Feel free to send me your CV and describe how you would like to contribute. If you did not receive a reponse from me in one week, feel free to send it again (I apologize in advance). Thanks! You could also find other faculty who work on related topics at Northeastern in the Statstical Machine Learning group and the Theory group.
Teaching
DS 5220, Supervised Machine Learning and Learning Theory, Fall 2021, Fall 2022.
CS 7140, Advanced Machine Learning, Spring 2021, Spring 2022.
CS 7180, Special Topics in AI: Algorithmic and Statistical Aspects of Deep Learning, Fall 2020.
Education/Experience
2020: postdoc at UPenn.
2019: Ph.D. from Stanford (advisors: Ashish Goel and Greg Valiant).
2013: Research assistant at Nanyang Technological University (advisor: Ning Chen).
2012: B.Eng. from Shanghai Jiao Tong University.