Speaker
Prof.
Kenji Harada
(Graduate School of Informatics, Kyoto University)
Description
We proposed the Adaptive Tensor Tree (ATT) method, which uses the tensor tree network within the Born machine framework to construct a generative model. This method expresses the target distribution function as the squared amplitude of a quantum wave function represented by a tensor tree. The core concept of the ATT method involves dynamically optimizing the tree structure to minimize the bond mutual information.
In this presentation, we introduce a new technique that utilizes an annealing process on mini-batch samples to enhance the performance of the ATT method. We will demonstrate the effectiveness of this new ATT approach using various datasets.
Primary author
Prof.
Kenji Harada
(Graduate School of Informatics, Kyoto University)
Co-authors
Prof.
Naoki Kawashima
(Institute for Solid State Physics, The University of Tokyo)
Prof.
Tsuyoshi Okubo
(Institute for Physics of Intelligence, The University of Tokyo)
