Skip to Main content Skip to Navigation
Conference papers

Sliced-Wasserstein normalizing flows: beyond maximum likelihood training

Abstract : Despite their advantages, normalizing flows generally suffer from several shortcomings including their tendency to generate unrealistic data (e.g., images) and their failing to detect out-of-distribution data. One reason for these deficiencies lies in the training strategy which traditionally exploits a maximum likelihood principle only. This paper proposes a new training paradigm based on a hybrid objective function combining the maximum likelihood principle (MLE) and a sliced-Wasserstein distance. Results obtained on synthetic toy examples and real image data sets show better generative abilities in terms of both likelihood and visual aspects of the generated samples. Reciprocally, the proposed approach leads to a lower likelihood of out-of-distribution data, demonstrating a greater data fidelity of the resulting flows.
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-03723029
Contributor : Nicolas Dobigeon Connect in order to contact the contributor
Submitted on : Wednesday, July 13, 2022 - 7:34:32 PM
Last modification on : Wednesday, September 7, 2022 - 8:14:05 AM

File

Coeurdoux_ESANN_2022.pdf
Files produced by the author(s)

Identifiers

Citation

Florentin Coeurdoux, Nicolas Dobigeon, Pierre Chainais. Sliced-Wasserstein normalizing flows: beyond maximum likelihood training. 30 th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2022), Oct 2022, Bruges, Belgium. ⟨10.48550/arXiv.2207.05468⟩. ⟨hal-03723029⟩

Share

Metrics

Record views

46

Files downloads

4