RIKEN Center for Advanced Intelligence Project Nonconvex Learning Theory Team
Team Leader: Takafumi Kanamori (Ph.D.)
In the nonconvex Learning Theory Team, we focus on developing machine learning algorithms and statistical theory to deal with massive high-dimensional data observed from heterogeneous environments. In particular, our team is intensively working on learning theory for multi-domain data, transfer learning, adversarial learning, statistical learning with divergence measures, information geometry, robust statistics, uncertain optimization. Furthermore, we study computationally efficient optimization methods for learning algorithms. Our aim is to establish theoretical foundations to design flexible learning methods that unify information gathered from heterogeneous data domains.
- Statistical Estimation for large-scale models using divergence measures
- Learning Theory of statistical inference with adversarial losses
- Extension of multimodal information integration and information-transfer learning
Main Research Fields
Related Research Fields
- Mathematical & Physical Sciences
- Statistical Science
- Applied Mathematics
- Mathematical Statistics
- Machine Learning
H. Sasaki, T Sakai, T. Kanamori.:
"Robust modal regression with direct gradient approximation of modal regression risk"
The Conference on Uncertainty in Artificial Intelligence (UAI2020). August 2020
M. Uehara, T. Kanamori, T. Takenouchi, T. Matsuda.:
"A Unified Statistically Efficient Estimation Framework for Unnormalized Models."
The 23rd International Conference on Artificial Intelligence and Statistics (AISTATS 2020).
S. Liu, T. Kanamori, W. Jitkrittum, Y. Chen.:
"Fisher Efficient Inference of Intractable Models."
The Neural Information Processing Systems (NeurIPS 2019)
K. Matsui, W. Kumagai, K. Kanamori, M. Nishikimi, T. Kanamori.:
"Variable Selection for Nonparametric Learning with Power Series Kernels."
Neural Computation, 31(8):1718-1750, August 2019.
W. Kumagai, T. Kanamori.:
"Risk Bound of Transfer Learning using Parametric Feature Mapping and Its Application to Sparse Coding."
Machine learning 108, pp. 1975--2008, May 2019.
H. Sasaki, T. Kanamori, A. Hyvarinen, and M. Sugiyama.:
"Mode-Seeking Clustering and Density Ridge Estimation via Direct Estimation of Density-Derivative-Ratios."
Journal of Machine Learning Research, Volume 18, Pages, 1--47, April, 2018.
T. Takenouchi, T. Kanamori.:
"Statistical Inference with Unnormalized Discrete Models and Localized Homogeneous Divergences."
Journal of Machine Learning Research, vol. 18, num. 56, pages 1--26, July 2017.
T. Kanamori, T. Takenouchi.:
"Graph-based Composite Local Bregman Divergences on Discrete Sample Spaces."
Neural Networks, Volume 95, Pages 44--56, November 2017.
T. Kanamori, S. Fujiwara, A. Takeda.:
"Robustness of Learning Algorithms using Hinge Loss with Outlier Indicators."
Neural Networks, Volume 94, Pages 173--191, October 2017.
K. Matsui, W. Kumagai, T. Kanamori.:
"Parallel Distributed Block Coordinate Descent Methods based on Pairwise Comparison Oracle."
Journal of Global Optimization, Volume 69, Issue 1, pp 1--21, September 2017.
- Takafumi Kanamori
- Team Leader
- Kosaku Takanashi
- Postdoctoral Researcher
|Seeking a Research Scientist or Postdoctoral Researcher (W20287)||Open until filled|
Department of Mathematical and Computing Science, Tokyo Institute of Technology,
Ookayama, Meguro-ku, Tokyo 152-8552 Japan
Email: kanamori [at] c.titech.ac.jp