• Corpus ID: 227347303

Using topological autoencoders as a filtering function for global and local topology

@article{Cornell2020UsingTA,
  title={Using topological autoencoders as a filtering function for global and local topology},
  author={Filip Cornell},
  journal={ArXiv},
  year={2020},
  volume={abs/2012.03383}
}
Choosing a suitable filtering function for the Mapper algorithm can be difficult due to its arbitrariness and domain-specific requirements. Finding a general filtering function that can be applied across domains is therefore of interest, since it would improve the representation of manifolds in higher dimensions. In this extended abstract, we propose that topological autoencoders is a suitable candidate for this and report initial results strengthening this hypothesis for one set of high… 
[PDF] Semantic Reader

Figures and Tables from this paper

References

SHOWING 1-8 OF 8 REFERENCES

Statistical Analysis and Parameter Selection for Mapper

It is shown that the Mapper is an optimal estimator of the Reeb graph, which gives, as a byproduct, a method to automatically tune its parameters and compute confidence regions on its topological features, such as its loops and flares.

Visualizing Data using t-SNE

A new technique called t-SNE that visualizes high-dimensional data by giving each datapoint a location in a two or three-dimensional map, a variation of Stochastic Neighbor Embedding that is much easier to optimize, and produces significantly better visualizations by reducing the tendency to crowd points together in the center of the map.

Convergence between Categorical Representations of Reeb Space and Mapper

Using tools from category theory, it is formally proved that the convergence between the Reeb space and mapper is proved in terms of an interleaving distance between their categorical representations.

A global geometric framework for nonlinear dimensionality reduction.

An approach to solving dimensionality reduction problems that uses easily measured local metric information to learn the underlying global geometry of a data set and efficiently computes a globally optimal solution, and is guaranteed to converge asymptotically to the true structure.

ShapeVis: High-dimensional Data Visualization at Scale

ShapeVis is presented, a scalable visualization technique for point cloud data inspired from topological data analysis that captures the underlying geometric and topological structure of the data in a compressed graphical representation and demonstrates how it captures the structural characteristics of real and synthetic data sets.

Topological Methods for the Analysis of High Dimensional Data Sets and 3D Object Recognition

The method, called Mapper, is based on the idea of partial clustering of the data guided by a set of functions defined on the data, and is not dependent on any particular clustering algorithm, i.e. any clustering algorithms may be used with Mapper.

Extracting insights from the shape of complex data using topology

The method combines the best features of existing standard methodologies such as principal component and cluster analyses to provide a geometric representation of complex data sets to find subgroups in data sets that traditional methodologies fail to find.

A Density Based Algorithm for Discovering Density Varied Clusters in Large Spatial Databases

This paper proposes a density varied DBSCAN algorithm which is capable to handle local density variation within the cluster and shows that the proposed clustering algorithm gives optimized results.