{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/latent-bernoulli-autoencoder","title":"Latent Bernoulli Autoencoder","arxiv_id":null,"date":"2020-01-01","proceeding":"ICML 2020 1","authors":["Jiri Fajtl","Vasileios Argyriou","Dorothy Monekosso","Paolo Remagnino"],"abstract":"In this work, we pose a question whether it is possible to design and train an\nautoencoder model in an end-to-end fashion to learn latent representations in\nmultivariate Bernoulli space, and achieve performance comparable with the\ncurrent state-of-the-art variational methods. Moreover, we investigate how to\ngenerate novel samples and perform smooth interpolation in the binary latent\nspace.  To meet our objective, we propose a simplified deterministic model\nwith a straight-through estimator to learn the binary latents and show its\ncompetitiveness with the latest VAE methods.  Furthermore, we propose a novel\nmethod based on a random hyperplane rounding for sampling and smooth\ninterpolation in the multivariate Bernoulli latent space.  Although not a main\nobjective, we demonstrate that our methods perform on par or better than the\ncurrent state-of-the-art methods on common CelebA, CIFAR-10 and MNIST \ndatasets. PyTorch code and trained models to reproduce published results \nwill be released with the camera ready version.","url_abs":"https://proceedings.icml.cc/static/paper_files/icml/2020/3022-Paper.pdf","url_pdf":"https://proceedings.icml.cc/static/paper_files/icml/2020/3022-Paper.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"latent-bernoulli-autoencoder","repo_url":"https://github.com/ok1zjf/lbae","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}