4959 2020 eng 1254 1261 8 325 conferenceobject 0 -- -- -- One of the major shortcomings of variational autoencoders is the inability to produce generations from the individual modalities of data originating from mixture distributions. This is primarily due to the use of a simple isotropic Gaussian as the prior for the latent code in the ancestral sampling procedure for the data generations. We propose a novel formulation of variational autoencoders, conditional prior VAE (CP-VAE), which learns to differentiate between the individual mixture components and therefore allows for generations from the distributional data clusters. We assume a two-level generative process with a continuous (Gaussian) latent variable sampled conditionally on a discrete (categorical) latent component. The new variational objective naturally couples the learning of the posterior and prior conditionals, and the learning of the latent categories encoding the multimodality of the original data in an unsupervised manner. The data-dependent conditional priors are then used to sample the continuous latent code when generating new samples from the individual mixture components corresponding to the multimodal structure of the original data. Our experimental results illustrate the generative performance of our new model comparing to multiple baselines. ECAI 2020 - 24th European Conference on Artificial Intelligence, 29 August-8 September 2020, Santiago de Compostela, Spain, August 29 - September 8, 2020 - Including 10th Conference on Prestigious Applications of Artificial Intelligence (PAIS 2020) 10.3233/FAIA200226 @inproceedingsDBLP:conf/ecai/LavdaGK20, author = Frantzeska Lavda and Magda Gregorová and Alexandros Kalousis, editor = Giuseppe De Giacomo and Alejandro Catalá and Bistra Dilkina and Michela Milano and Senén Barro and Alberto Bugarı́n and Jérôme Lang, title = Improving VAE Generations of Multimodal Data Through Data-Dependent Conditional Priors, booktitle = ECAI 2020 - 24th European Conference on Artificial Intelligence, 29 August-8 September 2020, Santiago de Compostela, Spain, August 29 - September 8, 2020 - Including 10th Conference on Prestigious Applications of Artificial Intelligence (PAIS 2020), series = Frontiers in Artificial Intelligence and Applications, volume = 325, pages = 1254–1261, publisher = IOS Press, year = 2020, url = https://doi.org/10.3233/FAIA200226, doi = 10.3233/FAIA200226, timestamp = Mon, 26 Jun 2023 01:00:00 +0200, biburl = https://dblp.org/rec/conf/ecai/LavdaGK20.bib, bibsource = dblp computer science bibliography, https://dblp.org md5:943301ce76def7f1ce2b939d56b515fb 2023-12-05T10:39:47+00:00 /tmp/phpbqftxu bibtex 656efdf3f01634.07906115 false true Frantzeska Lavda Magda Gregorová Alexandros Kalousis