Tongtong FANG | The Institute of Statistical Mathematics

About Me

I am a postdoctoral researcher at The Institute of Statistical Mathematics in Tokyo, working with Prof. Kenji Fukumizu. Before that, I received my Ph.D. degree from The University of Tokyo in 2024, under the supervision of Prof. Masashi Sugiyama.

My research primarily focuses on trustworthy machine learning, with the goal of developing, analyzing, and evaluating learning systems to ensure their safety, reliability, and robustness in real-world applications.

Research Interests

Transfer Learning: importance weighting, domain adaptation, representation learning

News

[Dec. 2023] Our paper about generalized importance weighting will be presented at NeurIPS’23 as a spotlight presentation.

Publications

Generalizing Importance Weighting to A Universal Solver for Distribution Shift Problems

Tongtong Fang, Nan Lu, Gang Niu, Masashi Sugiyama

Thirty-seventh Conference on Neural Information Processing Systems (NeurIPS 2023).

PDF Code Spotlight
Rethinking Importance Weighting for Deep Learning under Distribution shift

Tongtong Fang*, Nan Lu*, Gang Niu, Masashi Sugiyama (*Equal contribution)

Thirty-fourth Conference on Neural Information Processing Systems (NeurIPS 2020).

PDF Code Spotlight
Rethinking Importance Weighting for Transfer Learning

Nan Lu, Tianyi Zhang, Tongtong Fang, Takeshi Teshima, Masashi Sugiyama

Federated and Transfer Learning (pp. 185-231), Springer International Publishing, 2022.

PDF As a book chapter

Experience

2023.04 - 2025.04 JSPS Fellow (DC2, PD), The Japan Society for the Promotion of Science
2021.11 - 2024.08 Research Assistant, The University of Tokyo
2018.11 - 2019.07 Research Intern, RIKEN Center for Advanced Intelligence Project (AIP)

Services

Conference Reviewers

ICML

NeurIPS

ICLR

AISTATS

ACML

Journal Reviewers

Machine Learning Journal, Transactions on Machine Learning Research, Neural Networks, Neural Processing Letters