About me
Research interests
My central focus of research is in theoretical understanding of statistical machine learning, particularly from the following perspectives.
- Learning theory of loss functions: Loss functions are interesting because they characterize a large portion of task properties such as adversarial robustness (COLT2020) and imbalancedness (AISTATS2020, AISTATS2021). Some task properties can be attained simultaneously by proper losses (COLT2023).
- Evaluation metrics of predictions and representations. Recently, I am interested in how it is possible to learn good representations via similarity in light of a downstream task (ICML2018, AISTATS2022, ICML2022).
You may have a look at the slides of my past (and slightly outdated…) talks such as this to see my tastes.
News
🚨 リサーチアシスタント募集中
JSTさきがけ「未来数理科学」の研究課題「損失関数設計と最適化ダイナミクスの協調」に関連して、損失関数、最適化、ニューラルネットの帰納バイアス、表現学習に関する研究業務に携わっていただける方を募集しています。詳しくは直接ご連絡ください。
- Mar 16, 2025: 3月31日にナカニシヤ出版より共著「『事実』の交差点」が出版されます。異なる専門分野における「事実」の記述と形成に関して相互考察、議論を行った論文集です。私は機械学習の観点から「事実」の形成とその主観性、権力性について寄稿しました。ぜひ手にとっていただければと思います。
- Feb 25, 2025: I moved to the Institute of Statistical Mathematics as an associate professor.
- Jan 23, 2025: Our three new papers are accepted by AISTATS2025: (1) unified understanding of online inverse optimization via Fenchel-Young loss, (2) non-principal-centric model of inverse optimization via prediction market, and (3) a new loss class extending proper loss to incorporate the focal loss. Additionally, our two papers are accepted by ICLR2025: (1) hippocampus-inspired self-supervised learning and (2) scheduled knowledge distillation for language modeling.
- Nov 7, 2024: Our work on decentralized learning & Zipfian whitening got IEICE TC-IBISML Research Award and IBIS2024 Best Award, respectively!
- Sep 17, 2024: My grant proposal to JST-PRESTO (Japanese govermental 3.5 year research funding for individual researchers in natural science) has been accepted (official info).
- Jul 10, 2024: We are going to organize ECCV 2024 Workshop “Dark Side of GenAI and Beyond” at Milano, Italy, on Septempber 29. We highly encourage those who have ideas on possible social effects of generative/foundational models and their remedies (regardless of the maturity) to give a poster talk. The deadline is July 31.
- May 9, 2024: Our paper “Online Structured Prediction with Fenchel–Young Losses and Improved Surrogate Regret for Online Multiclass Classification with Logistic Loss” has been accepted by COLT2024.
- May 2, 2024: Our paper “Self-attention Networks Localize When QK-eigenspectrum Concentrates” has been accepted by ICML2024.
- Jan 20, 2024: Our paper “Fast 1-Wasserstein Distance Approximations Using Greedy Strategies” has been accepted by AISTATS2024.
Upcoming travels
- Apr 24-30: Singapore (ICLR etc.)
- May 2-5: Phuket (AISTATS)