3rd December
9:30 - 10:30
Nihat Ay, Max Planck Institute for Mathematics in the Sciences, Germany
Towards the Information Geometry of Embodied Agents
10:45 - 11:45
Takafumi Kanamori, Nagoya University, Japan
Legendre Transformation in Machine Learning
13:30 - 14:30
Mark Girolami, University of Warwick, UK
The Role of Information Geometry in Contemporary Computational Statistics
14:45 - 15:45
Shinto Eguchi, The Institute of Statistical Mathematics, Japan
Path Connectivity on a Probability Density Function Space
16:00 - 17:00
Fumiyasu Komaki, University of Tokyo, Japan
A Projection Method for Improving Prediction
4th December
9:30 - 10:30
Guido Montufar, Max Planck Institute for Mathematics in the Sciences, Germany
Information Divergence from Statistical Models Defined by Neural Networks
10:45 - 11:45
Shun-ichi Amari, RIKEN, Japan
Information Geometry of Bayesian Statistics, Restricted Boltzmann Machine Toward Understanding of Deep Learning
13:30 - 14:30
Frank Nielsen, Ecole Polytechnique, France
Fundamentals of Algorithms and Data-structures in Information-Geometric Spaces
14:45 - 15:45
Atsumi Ohara, University of Fukui, Japan
On Representing Functions of Probability Distributions and Conformal Flattening
16:00 - 17:00
Kenji Fukumizu, The Institute of Statistical Mathematics, Japan
Score Matching Estimation with Infinite Dimensional Exponential Family
5th December
9:30 - 10:30
Hiroshi Matsuzoe, Nagoya Institute of Technology, Japan
Geometry of Deformed Exponential Families
10:45 - 11:45
Masayuki Henmi, The Institute of Statistical Mathematics, Japan
Statistical Manifolds Admitting Torsion Induced from Estimating Functions
13:30 - 14:30
Takashi Tsuchiya, National Graduate Institute for Policy Research, Japan
von Neumann Entropy, Matrix Monotonicity and Polynomial-Time Interior-point Algorithms
14:45 - 15:45
Shiro Ikeda, The Institute of Statistical Mathematics, Japan
Optimization of Probability Measure and Information Geometry
|