Center for Advanced Intelligence Project

14 papers accepted at AISTATS 2026, a major international conference on machine learning.

AISTATS 2026

[Accepted Papers]

• Oral

• Gaussian Equivalence for Self-Attention: Asymptotic Spectral Analysis of Attention Matrix
  Tomohiro Hayase (CoeFont)
  Benoît Collins (Kyoto University)
  Ryo Karakida (AIST / RIKEN AIP)

• Spotlight

• Precise Dynamics of Diagonal Linear Networks: A Unifying Analysis by Dynamical Mean-Field Theory
  Sota Nishiyama (University of Tokyo / RIKEN AIP)
  Masaaki Imaizumi (University of Tokyo / RIKEN AIP)
• Tight Regret Upper and Lower Bounds for Optimistic Hedge in Two-Player Zero-Sum Games
  Taira Tsuchiya (University of Tokyo / RIKEN AIP)

• Poster

• Causal Additive Models with Unobserved Causal Paths and Backdoor Paths
  Thong Pham (Shiga University / RIKEN AIP)
  Takashi Nicholas Maeda (Gakushuin University / Shiga University / RIKEN AIP)
  Shohei Shimizu (University of Osaka / Shiga University / RIKEN AIP)
• Dualformer: Time-Frequency Dual Domain Learning for Long-term Time Series Forecasting
  Jingling Bai (University of Osaka / RIKEN AIP)
  Yoshinobu Kawahara (University of Osaka / RIKEN AIP)
• Generalized Correlation Shifting for Lasso
  Izuru Miyazaki (The Graduate University for Advanced Studies)
  Hironori Fujisawa (The Institute of Statistical Mathematics / The Graduate University for Advanced Studies / RIKEN AIP)
• Learning Linear Regression with Low-Rank Tasks in-Context
  Kaito Takanami (University of Tokyo / Ochanomizu University)
  Takashi Takahashi (University of Tokyo)
  Yoshiyuki Kabashima (University of Tokyo)
• Mixture Proportion Estimation and Weakly-supervised Kernel Test for Conditional Independence,
  Yushi Hirose (Science Tokyo)
  Akito Narahara (Science Tokyo)
  Takafumi Kanamori (Science Tokyo / RIKEN AIP)
• Neuron Block Dynamics for XOR Classification with Zero-Margin
  Guillaume Braun (RIKEN AIP)
  Masaaki Imaizumi (University of Tokyo / RIKEN AIP)
• Optimal Learning in Games under Delayed Feedback
  Ruotong Zhuang (University of Tokyo)
  Taira Tsuchiya (University of Tokyo / RIKEN AIP)
  Shinji Ito (University of Tokyo / RIKEN AIP)
• Provable Target Sample Complexity Improvements as Pre-Trained Models Scale
  Kazuto Fukuchi (University of Tsukuba / RIKEN AIP)
  Ryuichiro Hataya (SB Intuitions / Kyoto University)
  Kota Matsui (Kyoto University / Shiga University / Institute of Science Tokyo)
• Rate-optimal Design for Anytime Best Arm Identification
  Junpei Komiyama (MBZUAI / RIKEN AIP)
  Kyoungseok Jang (Chung-Ang University)
  Junya Honda (Kyoto University / RIKEN AIP)
• Towards Characterizing the Complexity of Riemannian Online Convex Optimization
  Hibiki Fukushima (University of Tokyo)
  Hiroshi Hirai (Nagoya University)
  Shinji Ito (University of Tokyo / RIKEN AIP)
• Tractable Shapley Values and Interactions via Tensor Networks
  Farzaneh Heidari (DIRO & Mila / Université de Montréal)
  Chao Li (RIKEN AIP)
  Guillaume Rabusseau (DIRO & Mila / Université de Montréal)

Related Laboratories