Witness Autoencoder: Shaping the Latent Space with Witness Complexes Download PDF

Oct 10, 2020 (edited Dec 02, 2020)NeurIPS 2020 Workshop TDA and Beyond Blind SubmissionReaders: Everyone
  • Keywords: Autoencoder, Witness Complex, Manifold Learning, Isometry
  • TL;DR: We present a Witness Autoencoder (W-AE) – an autoencoder that captures geodesic distances of the data in the latent space.
  • Abstract: We present a Witness Autoencoder (W-AE) – an autoencoder that captures geodesic distances of the data in the latent space. Our algorithm uses witness complexes to compute geodesic distance approximations on a mini-batch level, and leverages topological information from the entire dataset while performing batch-wise approximations. This way, our method allows to capture the global structure of the data even with a small batch size, which is beneficial for large-scale real-world data. We show that our method captures the structure of the manifold more accurately than the recently introduced topological autoencoder (TopoAE).
  • Previous Submission: No
  • Poster: pdf
1 Reply

Loading