張 一凡 · チョウ イーファン · [ʈʂāŋ īː fǽn]
yivanzhang@ms.k.u-tokyo.ac.jp
[CV] [Google Scholar] [DBLP] [GitHub]
Hi, this is Yivan! I’m a Ph.D. student supervised by Prof. Masashi Sugiyama at the University of Tokyo.
I’m interested in the theory and application of machine learning, especially those fundamental and foundational problems that can be instantiated in seemingly different fields.
I’m currently working on applied category theory in machine learning, including:
- supervision, representation, and generalization
- probability, uncertainty, and causality
- invariance and equivariance
- logic and metric
- discrete structures: sets, lists, trees, and graphs
I’m also interested in functional programming. I love things that were discovered.
My research is/was generously supported by RIKEN Center for Advanced Intelligence Project (RIKEN AIP), Japan Society for the Promotion of Science (JSPS), Japan Science and Technology Agency (JST), and Microsoft Research Asia (MSRA).
Publications
Conference
- A Category-theoretical Meta-analysis of Definitions of Disentanglement
Yivan Zhang, Masashi Sugiyama
International Conference on Machine Learning 2023 (ICML’23)
International Workshop on Symbolic-Neural Learning 2023 (SNL’23)
[arXiv] [ICML’23] [OpenReview] [slides] [poster] - Learning from Aggregate Observations
Yivan Zhang, Nontawat Charoenphakdee, Zhenguo Wu, Masashi Sugiyama
Neural Information Processing Systems 2020 (NeurIPS’20)
[arXiv] [NeurIPS’20] [code]
Workshop
- Enriching Disentanglement: Definitions to Metrics
Yivan Zhang, Masashi Sugiyama
Annual Workshop on Topology, Algebra, and Geometry in Machine Learning (TAG-ML) at International Conference on Machine Learning 2023 (ICML’23)
[arXiv] [TAG-ML’23] [poster]
Preprint
- Equivariant Disentangled Transformation for Domain Generalization under Combination Shift
Yivan Zhang, Jindong Wang, Xing Xie, Masashi Sugiyama
[arXiv]