Image of Jane Smith

Direct Ph.D. student
Computer Science (Informatik)

ETH Zürich
OAT, Andreasstrasse 5
8092 Zürich
chungentsai1007 [at] gmail [dot] com

Flag Counter

I am a first-year direct PhD student in Computer Science at ETH Zürich. I am fortunate to be advised by Prof. Niao He and to be a member of the ODI group, which is part of the Institute for Machine Learning. Previously, I obtained my bachelor's degree in Computer Science and Information Engineering at National Taiwan University, where I was advised by Prof. Yen-Huan Li.

Research

I work in optimization theory and machine learning theory, where I design algorithms and prove theoretical guarantees. I have served as a reviewer for NeurIPS 2024, ICLR 2025, and ICML 2025.

During my undergraduate studies, I focused on designing first-order algorithms for minimizing logarithmic losses. This problem arises in quantum state tomography and positron emission tomography, or more generally, in the Poisson inverse problem. The online counterpart of this problem is the online portfolio selection problem. We proved the first data-dependent bounds for online portfolio selection, which was published at NeurIPS 2023.

Selected Publications

Please check my Google scholar profile for the full list.

  • [ALT 2023] Online self-concordant and relatively smooth minimization, with applications to online portfolio selection and learning quantum states
    Chung-En Tsai, Hao-Chung Cheng, and Yen-Huan Li

  • [NeurIPS 2023] Data-dependent bounds for online portfolio selection without Lipschitzness and smoothness
    Chung-En Tsai, Ying-Ting Lin, and Yen-Huan Li

  • [AISTATS 2024] Fast minimization of expected logarithmic loss via stochastic dual averaging
    Chung-En Tsai, Hao-Chung Cheng, and Yen-Huan Li

Education

  • Direct Ph.D. in Computer Science, 2024-Present
    Eidgenössische Technische Hochschule Zürich (ETH Zürich)
  • B.Sc. in Computer Science and Information Engineering, 2019-2023
    National Taiwan University