{"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/bruno-a-deep-recurrent-model-for-exchangeable","title":"BRUNO: A Deep Recurrent Model for Exchangeable Data","arxiv_id":"1802.07535","date":"2018-02-21","proceeding":"NeurIPS 2018 12","authors":["Iryna Korshunova","Jonas Degrave","Ferenc Huszár","Yarin Gal","Arthur Gretton","Joni Dambre"],"abstract":"We present a novel model architecture which leverages deep learning tools to\nperform exact Bayesian inference on sets of high dimensional, complex\nobservations. Our model is provably exchangeable, meaning that the joint\ndistribution over observations is invariant under permutation: this property\nlies at the heart of Bayesian inference. The model does not require variational\napproximations to train, and new samples can be generated conditional on\nprevious samples, with cost linear in the size of the conditioning set. The\nadvantages of our architecture are demonstrated on learning tasks that require\ngeneralisation from short observed sequences while modelling sequence\nvariability, such as conditional image generation, few-shot learning, and\nanomaly detection.","url_abs":"http://arxiv.org/abs/1802.07535v3","url_pdf":"http://arxiv.org/pdf/1802.07535v3.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":"bruno-a-deep-recurrent-model-for-exchangeable","repo_url":"https://github.com/IraKorshunova/bruno","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"bruno-a-deep-recurrent-model-for-exchangeable","repo_url":"https://github.com/christabella/bruno","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"bruno-a-deep-recurrent-model-for-exchangeable","repo_url":"https://github.com/lupalab/flowscan","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"anomaly-detection","task_name":"Anomaly Detection"},{"task_slug":"bayesian-inference","task_name":"Bayesian Inference"},{"task_slug":"conditional-image-generation","task_name":"Conditional Image Generation"},{"task_slug":"few-shot-learning","task_name":"Few-Shot Learning"},{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"model","task_name":"model"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1802.07535","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}