Timezone: »
Spotlight
Legendre Decomposition for Tensors
Mahito Sugiyama · Hiroyuki Nakahara · Koji Tsuda
We present a novel nonnegative tensor decomposition method, called Legendre decomposition, which factorizes an input tensor into a multiplicative combination of parameters. Thanks to the well-developed theory of information geometry, the reconstructed tensor is unique and always minimizes the KL divergence from an input tensor. We empirically show that Legendre decomposition can more accurately reconstruct tensors than other nonnegative tensor decomposition methods.
Author Information
Mahito Sugiyama (National Institute of Informatics)
Hiroyuki Nakahara (RIKEN Brain Science Institute)
Koji Tsuda (The University of Tokyo / RIKEN)
Related Events (a corresponding poster, oral, or spotlight)
-
2018 Poster: Legendre Decomposition for Tensors »
Wed. Dec 5th through Thu the 6th Room Room 210 #92
More from the Same Authors
-
2021 Poster: Fast Tucker Rank Reduction for Non-Negative Tensors Using Mean-Field Approximation »
Kazu Ghalamkari · Mahito Sugiyama -
2020 : Contributed Talk 1: The Volume of Non-Restricted Boltzmann Machines and Their Double Descent Model Complexity »
Prasad Cheema · Mahito Sugiyama -
2020 : Convex Optimization for Blind Source Separation on a Statistical Manifold »
Simon Luo · lamiae azizi · Mahito Sugiyama -
2020 : Towards Geometric Understanding of Low-Rank Approximation »
Mahito Sugiyama · Kazu Ghalamkari -
2017 : ChemTS: An Efficient Python Library for De Novo Molecular Generation »
Koji Tsuda -
2017 Workshop: Machine Learning for Molecules and Materials »
Kristof Schütt · Klaus-Robert Müller · Anatole von Lilienfeld · José Miguel Hernández-Lobato · Klaus-Robert Müller · Alan Aspuru-Guzik · Bharath Ramsundar · Matt Kusner · Brooks Paige · Stefan Chmiela · Alexandre Tkatchenko · Anatole von Lilienfeld · Koji Tsuda -
2015 Poster: Halting in Random Walk Kernels »
Mahito Sugiyama · Karsten Borgwardt -
2013 Poster: Rapid Distance-Based Outlier Detection via Sampling »
Mahito Sugiyama · Karsten Borgwardt