Taiga Ishii
TL;DR
I'm Taiga Ishii (石井 太河)
D1 student at the University of Tokyo
Currently working on:
branching bias of texts and unsupervised parsing model
Email: taigarana [at] is.s.u-tokyo.ac.jp
Research Interest
The gap between linguistic and (purely) mathematical features of (natural) languages
Language universals, variations
Probing
Education
Doctor
2023/04 ~ Present: Department of Computer Science, The Graduate School of Information Science and Technology, The University of Tokyo
Master
2021/04 ~2023/03: Department of Computer Science, The Graduate School of Information Science and Technology, The University of Tokyo
Bachelor
2017/04 ~ 2021/03: Department of Information Science, School of Science, The University of Tokyo
Publications
International Conference (Refereed)
2023
Taiga Ishii & Yusuke Miyao. "Tree-shape Uncertainty for Analyzing the Inherent Branching Bias of Unsupervised Parsing Models". CoNLL 2023. [to appear]
Ryo Ueda, Taiga Ishii, & Yusuke Miyao. "On the Word Boundaries of Emergent Languages Based on Harris's Articulation Scheme". ICLR 2023. [paper]
International Conference (Non-refereed)
2022
Ryo Ueda, Taiga Ishii, Koki Washio, Yusuke Miyao. “Categorial Grammar Induction as a Compositionality Measure for Emergent Languages in Signaling Games.” Emergent Communication Workshop at the Tenth International Conference on Learning Representations (EmeCom at ICLR 2022). [paper]
Pre-print
2023
Taiga Ishii, Ryo Ueda, & Yusuke Miyao. "Empirical Analysis of the Inductive Bias of Recurrent Neural Networks by Discrete Fourier Transform of Output Sequences". ArXiv 2023. [paper]
Domestic Conference (Non-refereed)
Grants
JST SPRING GX: 2023/04~2026/03
Activities
2021/03 ~ Present: Contributing to MioGatto: Math Identifier-oriented Grounding Annotation Tool [URL]
2022/03 ~2023/03: Working as a researcher in CREST [数理的情報活用基盤] 令和3年度採択課題 「自然言語の非線形性の計算論モデル」
Hobby
Fishing
Anime
Tasting (Tea, Coffee, Alcohol, etc.)
Martial Arts