Search Results for author: Martin Royer

Found 5 papers, 3 papers with code

ATOL: Measure Vectorization for Automatic Topologically-Oriented Learning

no code implementations30 Sep 2019 Martin Royer, Frédéric Chazal, Clément Levrard, Umeda Yuhei, Ike Yuichi

Robust topological information commonly comes in the form of a set of persistence diagrams, finite measures that are in nature uneasy to affix to generic machine learning frameworks.

BIG-bench Machine Learning Time Series Analysis +1

PersLay: A Neural Network Layer for Persistence Diagrams and New Graph Topological Signatures

1 code implementation20 Apr 2019 Mathieu Carrière, Frédéric Chazal, Yuichi Ike, Théo Lacombe, Martin Royer, Yuhei Umeda

Persistence diagrams, the most common descriptors of Topological Data Analysis, encode topological properties of data and have already proved pivotal in many different applications of data science.

Graph Classification Topological Data Analysis

Adaptive Clustering through Semidefinite Programming

no code implementations NeurIPS 2017 Martin Royer

We analyze the clustering problem through a flexible probabilistic model that aims to identify an optimal partition on the sample X1,..., Xn.

Clustering

PECOK: a convex optimization approach to variable clustering

1 code implementation16 Jun 2016 Florentina Bunea, Christophe Giraud, Martin Royer, Nicolas Verzelen

The problem of variable clustering is that of grouping similar components of a $p$-dimensional vector $X=(X_{1},\ldots, X_{p})$, and estimating these groups from $n$ independent copies of $X$.

Statistics Theory Statistics Theory

Model Assisted Variable Clustering: Minimax-optimal Recovery and Algorithms

1 code implementation8 Aug 2015 Florentina Bunea, Christophe Giraud, Xi Luo, Martin Royer, Nicolas Verzelen

We quantify the difficulty of clustering data generated from a G-block covariance model in terms of cluster proximity, measured with respect to two related, but different, cluster separation metrics.

Clustering

Cannot find the paper you are looking for? You can Submit a new open access paper.