Nonparametric Generative Modeling with Conditional Sliced-Wasserstein Flows

Chao Du, Tianbo Li, Tianyu Pang, Shuicheng Yan, Min Lin
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:8565-8584, 2023.

Abstract

Sliced-Wasserstein Flow (SWF) is a promising approach to nonparametric generative modeling but has not been widely adopted due to its suboptimal generative quality and lack of conditional modeling capabilities. In this work, we make two major contributions to bridging this gap. First, based on a pleasant observation that (under certain conditions) the SWF of joint distributions coincides with those of conditional distributions, we propose Conditional Sliced-Wasserstein Flow (CSWF), a simple yet effective extension of SWF that enables nonparametric conditional modeling. Second, we introduce appropriate inductive biases of images into SWF with two techniques inspired by local connectivity and multiscale representation in vision research, which greatly improve the efficiency and quality of modeling images. With all the improvements, we achieve generative performance comparable with many deep parametric generative models on both conditional and unconditional tasks in a purely nonparametric fashion, demonstrating its great potential.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-du23c, title = {Nonparametric Generative Modeling with Conditional Sliced-{W}asserstein Flows}, author = {Du, Chao and Li, Tianbo and Pang, Tianyu and Yan, Shuicheng and Lin, Min}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {8565--8584}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/du23c/du23c.pdf}, url = {https://proceedings.mlr.press/v202/du23c.html}, abstract = {Sliced-Wasserstein Flow (SWF) is a promising approach to nonparametric generative modeling but has not been widely adopted due to its suboptimal generative quality and lack of conditional modeling capabilities. In this work, we make two major contributions to bridging this gap. First, based on a pleasant observation that (under certain conditions) the SWF of joint distributions coincides with those of conditional distributions, we propose Conditional Sliced-Wasserstein Flow (CSWF), a simple yet effective extension of SWF that enables nonparametric conditional modeling. Second, we introduce appropriate inductive biases of images into SWF with two techniques inspired by local connectivity and multiscale representation in vision research, which greatly improve the efficiency and quality of modeling images. With all the improvements, we achieve generative performance comparable with many deep parametric generative models on both conditional and unconditional tasks in a purely nonparametric fashion, demonstrating its great potential.} }
Endnote
%0 Conference Paper %T Nonparametric Generative Modeling with Conditional Sliced-Wasserstein Flows %A Chao Du %A Tianbo Li %A Tianyu Pang %A Shuicheng Yan %A Min Lin %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-du23c %I PMLR %P 8565--8584 %U https://proceedings.mlr.press/v202/du23c.html %V 202 %X Sliced-Wasserstein Flow (SWF) is a promising approach to nonparametric generative modeling but has not been widely adopted due to its suboptimal generative quality and lack of conditional modeling capabilities. In this work, we make two major contributions to bridging this gap. First, based on a pleasant observation that (under certain conditions) the SWF of joint distributions coincides with those of conditional distributions, we propose Conditional Sliced-Wasserstein Flow (CSWF), a simple yet effective extension of SWF that enables nonparametric conditional modeling. Second, we introduce appropriate inductive biases of images into SWF with two techniques inspired by local connectivity and multiscale representation in vision research, which greatly improve the efficiency and quality of modeling images. With all the improvements, we achieve generative performance comparable with many deep parametric generative models on both conditional and unconditional tasks in a purely nonparametric fashion, demonstrating its great potential.
APA
Du, C., Li, T., Pang, T., Yan, S. & Lin, M.. (2023). Nonparametric Generative Modeling with Conditional Sliced-Wasserstein Flows. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:8565-8584 Available from https://proceedings.mlr.press/v202/du23c.html.

Related Material