no code implementations • 1 Feb 2024 • Kurt Pasque, Christopher Teska, Ruriko Yoshida, Keiji Miura, Jefferson Huang
We introduce a simple, easy to implement, and computationally efficient tropical convolutional neural network architecture that is robust against adversarial attacks.
1 code implementation • 3 Sep 2023 • David Barnhill, Ruriko Yoshida, Georgios Aliatimis, Keiji Miura
In the last decade, developments in tropical geometry have provided a number of uses directly applicable to problems in statistical learning.
no code implementations • 9 Jun 2022 • Ruriko Yoshida, David Barnhill, Keiji Miura, Daniel Howe
In order to discover ``outlying'' gene trees which do not follow the ``main distribution(s)'' of trees, we propose to apply the ``tropical metric'' with the max-plus algebra from tropical geometry to a non-parametric estimation of gene trees over the space of phylogenetic trees.
no code implementations • 27 Jan 2021 • Ruriko Yoshida, Misaki Takamori, Hideyuki Matsumoto, Keiji Miura
Similar to SVMs, tropical SVMs classify data points using a tropical hyperplane under the tropical metric with the max-plus algebra.