The mantis repository contains the code corresponding to the paper:

Anirvan Sengupta, Mariano Tepper, Cengiz Pehlevan, Alexander Genkin, Dmitri Chklovskii
Manifold-tiling Localized Receptive Fields are Optimal in Similarity-preserving Neural Networks
NIPS 2018
https://www.biorxiv.org/content/early/2018/06/04/338947

Dependencies: cvxpy matplotlib numpy pillow scipy seaborn

Copyright (c) 2017 by Mariano Tepper, Flatiron Institute, Simons Foundation