Ph.D. Candidate @ Tsinghua CS
|
liujiashuo77@gmail.com
About Me
Hi, Im Jiashuo! I'm a fourth-year Ph.D. candidate in the Department of Computer Science & Technology at Tsinghua University, advised by Prof. Peng Cui.
I am currently a visiting student researcher at Prof. Jose H. Blanchet's group in the Department of Management Science & Engineering at Stanford University.
I work remotely with Prof. Hongseok Namkoong in the Decision, Risk, and Operations division at Columbia Business School.
I am also lucky to work with Prof. Bo Li.
My CV is here: (English/ Chinese).
My research lies at the interface of machine learning and operations research, including:
- Generalization under Distribution Shifts: Theoretical foundations and tools for understanding real-world distribution shifts;
- Data Heterogeneity: Algorithms to model and exploit data heterogeneity under distribution shifts;
- Distributionally robust optimization: Addressing the over-pessimism problem in real applications.
Recently, I'm focusing on the evaluation of LLMs as well as incorporating LLMs into robust optimization.
Invited Tutorials / Talks
- [3]Stability Evaluation via Distributional Perturbation Analysis (coming soon)
Invited presentation at 2024 INFORMS Annual Meeting
Session: Advances in Data-Driven Distributionally Robust Optimization
Seattle, U.S. - [2]Model the Data Heterogeneity for Out-of-Distribution Generalization
Tutorial at SIAM International Conference on Data Mining 2024 (SDM24)
Houston, TX, U.S - [1]Modeling & Exploiting Data Heterogeneity under Distribution Shifts
[slides] [video] Tutorial at Conference on Neural Information Processing Systems 2023 (NeurIPS 2023)
New Orleans, LA, U.S.
Selected as Favorite Papers/Presentations (9/3500+) by Two Sigma
Selected Papers
The full list of publications can be found in the publications page.* indicates equal contributions, † denotes alphabetical order.
Topic 1: Generalization under Distribution Shifts
- [5]Stability Evaluation via Distributional Perturbation Analysis
[paper] - [3]Towards Robust Out-of-Distribution Generalization Bounds via Sharpness
ICLR 2024, the 12th International Conference on Learning Representations.
Spotlight presentation (Top 5%).[paper] - [2]On the Need for a Language Describing Distribution Shifts: Illustrations on Tabular Datasets
NeurIPS 2023, Datasets and Benchmarks Track, the 37th Conference on Neural Information Processing Systems.
Selected as Favorite Papers/Presentations (9/3500+) by Two Sigma[paper] [GitHub]
Journal version under preparation
Topic 2: Data Heterogeneity
Topic 3: Distributionally Robust Optimization
- [4]Geometry-Calibrated DRO: Combating Over-Pessimism with Free Energy Implications
NeurIPS DistShift 2023, in NeurIPS 2023 Workshop on Distribution Shifts.
[workshop paper] [full paper] - [3]Distributionally Robust Learning with Stable Adversarial Training
IEEE TKDE 2022, IEEE Transactions on Knowledge and Data Engineering.
[paper] - [2]Distributionally Robust Optimization with Data Geometry
NeurIPS 2022, the 36th Conference on Neural Information Processing Systems.
Spotlight presentation (Top 5.2%).[paper]
2024
- [18]Rethinking the Evaluation Protocol of Domain Generalization
CVPR 2024, the Conference on Computer Vision and Pattern Recognition 2024
[paper] - [17]Distributionally Generative Augmentation for Fair Facial Attribute Classification
CVPR 2024, the Conference on Computer Vision and Pattern Recognition 2024 - [15]Towards Robust Out-of-Distribution Generalization Bounds via Sharpness
ICLR 2024, the 12th International Conference on Learning Representations.
Spotlight presentation (Top 5%).[paper]
2023
- [14]Geometry-Calibrated DRO: Combating Over-Pessimism with Free Energy Implications
NeurIPS 2023 DistShift Workshop, in NeurIPS 2023 Workshop on Distribution Shifts.
[workshop paper] [full paper] - [13]On the Need for a Language Describing Distribution Shifts: Illustrations on Tabular Datasets
NeurIPS 2023, Datasets and Benchmarks Track
the 37th Conference on Neural Information Processing Systems.
Selected as Favorite Papers/Presentations (9/3500+) by Two Sigma[paper][code] [python package] [downloads>2.6k] - [12]Offline Policy Evaluation in Large Action Spaces via Outcome-Oriented Action Grouping
WWW 2023, the ACM Web Conference 2023.[paper] - [11]Measure the Predictive Heterogeneity
ICLR 2023, the 11th International Conference on Learning Representations.
[paper]
2022
- [10]Distributionally Robust Learning with Stable Adversarial Training
IEEE TKDE 2022, IEEE Transactions on Knowledge and Data Engineering.
[paper] - [9]Distributionally Robust Optimization with Data Geometry
NeurIPS 2022, the 36th Conference on Neural Information Processing Systems.
Spotlight presentation (Top 5.2%).[paper] - [8]Towards the ultimate PMT waveform analysis for neutrino and dark matter experiments
JINST 2022, the Journal of Instrumentation.
[paper] - [7]Invariant Preference Learning for General Debiasing in Recommendation
KDD 2022, the SIGKDD Conference on Knowledge Discovery and Data Mining.
[paper]
2021
- [6]Triple Generative Adversarial Networks
TPAMI 2021, the Transactions on Pattern Analysis and Machine Intelligence.
[paper] - [2]Signed Graph Neural Network with Latent Groups
KDD 2021, the SIGKDD Conference on Knowledge Discovery and Data Mining.
[paper]
2020
- [1]Stable Learning via Differentiated Variable Decorrelation
KDD 2020, the SIGKDD Conference on Knowledge Discovery and Data Mining.
[paper]
Professional Services
Journal reviewer: Operations Research, IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), IEEE Transactions on Knowledge and Data Engineering (TKDE), IEEE Transactions on Multimedia (TMM)
Conference reviewer / Program committee: ICLR (2024), NeurIPS (2023), ICML (2022,2023,2024), UAI (2022,2023,2024), AAAI (2022), IJCAI (2022,2023), CVPR (2022,2023, 2024), ECCV (2024), ICCV (2023), CoLLAs (2022,2023,2024), AISTATS (2021,2023,2024)
Workshop reviewer / Program committee: NeurIPS DistShift (2023)
Awards and Fellowships
- Tsinghua-Huawei Scholarship (2023)
- Excellent Comprehensive Scholarship of Tsinghua University (for PhD. Student, 2022)
- National Scholarship for Ph.D. (2021)
- Apple Scholars in AI/ML Nomination (2021)
- Excellent Undergraduate, Tsinghua University (2020, 10%)
- TP-Link Scholarship (2019)
- Toyota Scholarship (2018)
- Excellent Academic Scholarship of Tsinghua University (2018, 2019)
- Excellent Comprehensive Scholarship of Tsinghua University (2017, 5%)
- Second-Class Freshmen Scholarship of Tsinghua University (2016~2019)
Teaching
- Software Engineering (Fall 2019, 2020, 2021, 2022, Spring 2022, 2023 TA)
- Object-oriented Programming (Summer 2022, TA)
Acknowledgements: based on the al-folio template by Maruan Al-Shedivat and Jiaming Song