{"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/causal-effect-inference-with-deep-latent","title":"Causal Effect Inference with Deep Latent-Variable Models","arxiv_id":"1705.08821","date":"2017-05-24","proceeding":"NeurIPS 2017 12","authors":["Christos Louizos","Uri Shalit","Joris Mooij","David Sontag","Richard Zemel","Max Welling"],"abstract":"Learning individual-level causal effects from observational data, such as\ninferring the most effective medication for a specific patient, is a problem of\ngrowing importance for policy makers. The most important aspect of inferring\ncausal effects from observational data is the handling of confounders, factors\nthat affect both an intervention and its outcome. A carefully designed\nobservational study attempts to measure all important confounders. However,\neven if one does not have direct access to all confounders, there may exist\nnoisy and uncertain measurement of proxies for confounders. We build on recent\nadvances in latent variable modeling to simultaneously estimate the unknown\nlatent space summarizing the confounders and the causal effect. Our method is\nbased on Variational Autoencoders (VAE) which follow the causal structure of\ninference with proxies. We show our method is significantly more robust than\nexisting methods, and matches the state-of-the-art on previous benchmarks\nfocused on individual treatment effects.","url_abs":"http://arxiv.org/abs/1705.08821v2","url_pdf":"http://arxiv.org/pdf/1705.08821v2.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":"causal-effect-inference-with-deep-latent","repo_url":"https://github.com/AMLab-Amsterdam/CEVAE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"causal-effect-inference-with-deep-latent","repo_url":"https://github.com/MorningBooks/A_oveview_of_awesome_causality_dataset","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"causal-effect-inference-with-deep-latent","repo_url":"https://github.com/MorningBooks/Causality","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"causal-effect-inference-with-deep-latent","repo_url":"https://github.com/belaalb/CEVAE-VampPrior","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"causal-effect-inference-with-deep-latent","repo_url":"https://github.com/rik-helwegen/cevae_pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"causal-effect-inference-with-deep-latent","repo_url":"https://github.com/uber/causalml/blob/master/causalml/inference/nn/cevae.py","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"causal-inference","task_name":"Causal Inference"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/causal-inference-on-ihdp","task":"Causal Inference","dataset":"IHDP","model":"CEVAE","rank_in_archive_order":10,"of":13,"metrics":{"Average Treatment Effect Error":"0.46"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1705.08821","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}