Reconstructing linearly embedded graphs: A first step to stratified space learning

In this paper, we consider the simplest class of stratified spaces -- linearly embedded graphs. We present an algorithm that learns abstract structure graph and models specific embedding from a point cloud sampled it. use tools inspiration computational geometry, algebraic topology, topological data analysis prove correctness identified under assumptions on embedding. The is implemented in Julia package http://github.com/yossibokor/Skyler.jl , which used for numerical simulations paper.