no code implementations • NeurIPS 2021 • Sebastian Zeng, Florian Graf, Christoph Hofer, Roland Kwitt
The problem of (point) forecasting $ \textit{univariate} $ time series is considered.
1 code implementation • 21 Jun 2019 • Christoph Hofer, Roland Kwitt, Mandar Dixit, Marc Niethammer
In particular, we control the connectivity of an autoencoder's latent space via a novel type of loss, operating on information from persistent homology.
4 code implementations • NeurIPS 2017 • Christoph Hofer, Roland Kwitt, Marc Niethammer, Andreas Uhl
Inferring topological and geometrical information from data can offer an alternative perspective on machine learning problems.