{"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/bayesian-learning-of-neural-network","title":"Bayesian Learning of Neural Network Architectures","arxiv_id":"1901.04436","date":"2019-01-14","proceeding":null,"authors":["Georgi Dikov","Patrick van der Smagt","Justin Bayer"],"abstract":"In this paper we propose a Bayesian method for estimating architectural\nparameters of neural networks, namely layer size and network depth. We do this\nby learning concrete distributions over these parameters. Our results show that\nregular networks with a learnt structure can generalise better on small\ndatasets, while fully stochastic networks can be more robust to parameter\ninitialisation. The proposed method relies on standard neural variational\nlearning and, unlike randomised architecture search, does not require a\nretraining of the model, thus keeping the computational overhead at minimum.","url_abs":"http://arxiv.org/abs/1901.04436v2","url_pdf":"http://arxiv.org/pdf/1901.04436v2.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":"bayesian-learning-of-neural-network","repo_url":"https://github.com/antonFJohansson/Bayesian-Learning-of-Neural-Network-Architectures","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"architecture-search","task_name":"Neural Architecture Search"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1901.04436","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}