Source author record

Hiroaki Kurihara

Explore this author’s publications, collaborators and research topics. This catalog record has not been claimed by its author.

ResearcherUnclaimed source record

Computational Geometry Machine Learning math.GT

Catalog footprint

Research overview

2works

3topics

4close collaborators

Actions

Is this your work?

Open graph Browse papers

preprint2022arXiv

On invariants of surfaces in the $3$ -sphere

In this paper we study isotopy classes of closed connected orientable surfaces in the standard $3$ -sphere. Such a surface splits the $3$ -sphere into two compact connected submanifolds, and by using their Heegaard splittings, we obtain a $2$ -component handlebody-link. In this paper, we first show that the equivalence class of such a 2-component handlebody-link up to attaching trivial $1$ -handles can recover the original surface. Therefore, we can reduce the study of surfaces in the $3$ -sphere to that of $2$ -component handlebody-links up to stabilizations. Then, by using $G$ -families of quandles, we construct invariants of $2$ -component handlebody-links up to attaching trivial $1$ -handles, which lead to invariants of surfaces in the $3$ -sphere. In order to see the effectiveness of our invariants, we will also show that our invariants can distinguish certain explicit surfaces in the $3$ -sphere.

preprint2022arXiv

RipsNet: a general architecture for fast and robust estimation of the persistent homology of point clouds

The use of topological descriptors in modern machine learning applications, such as Persistence Diagrams (PDs) arising from Topological Data Analysis (TDA), has shown great potential in various domains. However, their practical use in applications is often hindered by two major limitations: the computational complexity required to compute such descriptors exactly, and their sensitivity to even low-level proportions of outliers. In this work, we propose to bypass these two burdens in a data-driven setting by entrusting the estimation of (vectorization of) PDs built on top of point clouds to a neural network architecture that we call RipsNet. Once trained on a given data set, RipsNet can estimate topological descriptors on test data very efficiently with generalization capacity. Furthermore, we prove that RipsNet is robust to input perturbations in terms of the 1-Wasserstein distance, a major improvement over the standard computation of PDs that only enjoys Hausdorff stability, yielding RipsNet to substantially outperform exactly-computed PDs in noisy settings. We showcase the use of RipsNet on both synthetic and real-world data. Our open-source implementation is publicly available at https://github.com/hensel-f/ripsnet and will be included in the Gudhi library.

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Hiroaki Kurihara

Research overview

Is this your work?

2 publications

On invariants of surfaces in the $3$ -sphere

RipsNet: a general architecture for fast and robust estimation of the persistent homology of point clouds

See the researcher in context

Building this map preview

Hiroaki Kurihara

Research overview

Is this your work?

2 publications

On invariants of surfaces in the 333-sphere

RipsNet: a general architecture for fast and robust estimation of the persistent homology of point clouds

See the researcher in context

Building this map preview

On invariants of surfaces in the $3$ -sphere