Folded Hamiltonian Monte Carlo for Bayesian Generative Adversarial Networks

Narges Pourshahrokhi, Yunpeng Li, Samaneh Kouchaki, Payam Barnaghi
Proceedings of the 15th Asian Conference on Machine Learning, PMLR 222:1103-1118, 2024.

Abstract

Probabilistic modelling on Generative Adversarial Networks (GANs) within the Bayesian framework has shown success in estimating the complex distribution in literature. In this paper, we develop a Bayesian formulation for unsupervised and semi-supervised GAN learning. Specifically, we propose Folded Hamiltonian Monte Carlo (F-HMC) methods within this framework to learn the distributions over the parameters of the generators and discriminators. We show that the F-HMC efficiently approximates multi-modal and high dimensional data when combined with Bayesian GANs. Its composition improves run time and test error in generating diverse samples. Experimental results with high-dimensional synthetic multi-modal data and natural image benchmarks, including CIFAR-10, SVHN and ImageNet, show that F-HMC outperforms the state-of-the-art methods in terms of test error, run times per epoch, inception score and Frechet Inception Distance scores.

Cite this Paper

BibTeX
@InProceedings{pmlr-v222-pourshahrokhi24a, title = {{F}olded {H}amiltonian {M}onte {C}arlo for {B}ayesian {G}enerative {A}dversarial {N}etworks}, author = {Pourshahrokhi, Narges and Li, Yunpeng and Kouchaki, Samaneh and Barnaghi, Payam}, booktitle = {Proceedings of the 15th Asian Conference on Machine Learning}, pages = {1103--1118}, year = {2024}, editor = {Yanıkoğlu, Berrin and Buntine, Wray}, volume = {222}, series = {Proceedings of Machine Learning Research}, month = {11--14 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v222/pourshahrokhi24a/pourshahrokhi24a.pdf}, url = {https://proceedings.mlr.press/v222/pourshahrokhi24a.html}, abstract = {Probabilistic modelling on Generative Adversarial Networks (GANs) within the Bayesian framework has shown success in estimating the complex distribution in literature. In this paper, we develop a Bayesian formulation for unsupervised and semi-supervised GAN learning. Specifically, we propose Folded Hamiltonian Monte Carlo (F-HMC) methods within this framework to learn the distributions over the parameters of the generators and discriminators. We show that the F-HMC efficiently approximates multi-modal and high dimensional data when combined with Bayesian GANs. Its composition improves run time and test error in generating diverse samples. Experimental results with high-dimensional synthetic multi-modal data and natural image benchmarks, including CIFAR-10, SVHN and ImageNet, show that F-HMC outperforms the state-of-the-art methods in terms of test error, run times per epoch, inception score and Frechet Inception Distance scores.} }
Endnote
%0 Conference Paper %T Folded Hamiltonian Monte Carlo for Bayesian Generative Adversarial Networks %A Narges Pourshahrokhi %A Yunpeng Li %A Samaneh Kouchaki %A Payam Barnaghi %B Proceedings of the 15th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Berrin Yanıkoğlu %E Wray Buntine %F pmlr-v222-pourshahrokhi24a %I PMLR %P 1103--1118 %U https://proceedings.mlr.press/v222/pourshahrokhi24a.html %V 222 %X Probabilistic modelling on Generative Adversarial Networks (GANs) within the Bayesian framework has shown success in estimating the complex distribution in literature. In this paper, we develop a Bayesian formulation for unsupervised and semi-supervised GAN learning. Specifically, we propose Folded Hamiltonian Monte Carlo (F-HMC) methods within this framework to learn the distributions over the parameters of the generators and discriminators. We show that the F-HMC efficiently approximates multi-modal and high dimensional data when combined with Bayesian GANs. Its composition improves run time and test error in generating diverse samples. Experimental results with high-dimensional synthetic multi-modal data and natural image benchmarks, including CIFAR-10, SVHN and ImageNet, show that F-HMC outperforms the state-of-the-art methods in terms of test error, run times per epoch, inception score and Frechet Inception Distance scores.
APA
Pourshahrokhi, N., Li, Y., Kouchaki, S. & Barnaghi, P.. (2024). Folded Hamiltonian Monte Carlo for Bayesian Generative Adversarial Networks. Proceedings of the 15th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 222:1103-1118 Available from https://proceedings.mlr.press/v222/pourshahrokhi24a.html.

Related Material