Sylvester Normalizing Flows for Variational Inference

Rianne van den Berg, Leonard Hasenclever, Jakub Tomczak, Max Welling
Proceedings of the 34th Conference on Uncertainty in Artificial Intelligence, PMLR R16:392-401, 2018.

Abstract

Variational inference relies on flexible approximate posterior distributions. Normalizing flows provide a general recipe to construct flexible variational posteriors. We introduce Sylvester normalizing flows, which can be seen as a generalization of planar flows. Sylvester normalizing flows remove the well-known single-unit bottleneck from planar flows, making a single transformation much more flexible. We compare the performance of Sylvester normalizing flows against planar flows and inverse autoregressive flows and demonstrate that they compare favorably on several datasets.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR16-berg18a, title = {Sylvester Normalizing Flows for Variational Inference}, author = {Berg, Rianne van den and Hasenclever, Leonard and Tomczak, Jakub and Welling, Max}, booktitle = {Proceedings of the 34th Conference on Uncertainty in Artificial Intelligence}, pages = {392--401}, year = {2018}, editor = {Globerson, Amir and Silva, Ricardo}, volume = {R16}, series = {Proceedings of Machine Learning Research}, month = {06--10 Aug}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r16/main/assets/berg18a/berg18a.pdf}, url = {https://proceedings.mlr.press/r16/berg18a.html}, abstract = {Variational inference relies on flexible approximate posterior distributions. Normalizing flows provide a general recipe to construct flexible variational posteriors. We introduce Sylvester normalizing flows, which can be seen as a generalization of planar flows. Sylvester normalizing flows remove the well-known single-unit bottleneck from planar flows, making a single transformation much more flexible. We compare the performance of Sylvester normalizing flows against planar flows and inverse autoregressive flows and demonstrate that they compare favorably on several datasets.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Sylvester Normalizing Flows for Variational Inference %A Rianne van den Berg %A Leonard Hasenclever %A Jakub Tomczak %A Max Welling %B Proceedings of the 34th Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2018 %E Amir Globerson %E Ricardo Silva %F pmlr-vR16-berg18a %I PMLR %P 392--401 %U https://proceedings.mlr.press/r16/berg18a.html %V R16 %X Variational inference relies on flexible approximate posterior distributions. Normalizing flows provide a general recipe to construct flexible variational posteriors. We introduce Sylvester normalizing flows, which can be seen as a generalization of planar flows. Sylvester normalizing flows remove the well-known single-unit bottleneck from planar flows, making a single transformation much more flexible. We compare the performance of Sylvester normalizing flows against planar flows and inverse autoregressive flows and demonstrate that they compare favorably on several datasets. %Z Reissued by PMLR on 04 October 2026.
APA
Berg, R.v.d., Hasenclever, L., Tomczak, J. & Welling, M.. (2018). Sylvester Normalizing Flows for Variational Inference. Proceedings of the 34th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R16:392-401 Available from https://proceedings.mlr.press/r16/berg18a.html. Reissued by PMLR on 04 October 2026.

Related Material