Deep Invertible Autoencoders for Dimensionality Reduction of Dynamical Systems
An invertible AE architecture, named inv-AE, is proposed that computationally improves upon the stagnation of the reconstruction error typical of traditional AE architectures and can be combined with popular autoencoder-based ROM approaches, e.g., DL-ROM, to improve their accuracy.
Nicolò Botteghi, S. Glas, Christoph Brune
· Computer Methods in Applied... · 1 citation