[{"volume":37,"researchdata_availability":"yes","publication_identifier":{"eissn":["1049-5258"]},"intvolume":"        37","quality_controlled":"1","_id":"19509","department":[{"_id":"SyCr"},{"_id":"FrLo"}],"alternative_title":["Advances in Neural Information Processing Systems"],"acknowledgement":"We thank Piersilvio De Bartolomeis, and the full Causal Learning and Artificial Intelligence (CLAI) group at ISTA for the extremely helpful discussions. Riccardo Cadei was supported by a Google Research Scholar Award and a Google Initiated Gift to Francesco Locatello. We thank the Social Immunity team at ISTA, particularly Michaela Hönigsberger and Wilfrid Jean Louis, for supporting the ecological experiment and Farnaz Beikzadeh Abbasi, Luisa Fiebig and Martin Estermann for annotating ant behavior in ISTAnt.","conference":{"location":"Vancouver, Canada","end_date":"2024-12-15","name":"NeurIPS: Neural Information Processing Systems","start_date":"2024-12-09"},"external_id":{"arxiv":["2405.17151"]},"month":"12","oa_version":"Preprint","publisher":"Neural Information Processing Systems Foundation","das_tickbox":"1","doi":"10.52202/079017-0820","date_published":"2024-12-20T00:00:00Z","day":"20","abstract":[{"lang":"eng","text":"Machine Learning and AI have the potential to transform data-driven scientific discovery,\r\nenabling accurate predictions for several scientific phenomena. As many scientific questions\r\nare inherently causal, this paper looks at the causal inference task of treatment effect\r\nestimation, where the outcome of interest is recorded in high-dimensional observations\r\nin a Randomized Controlled Trial (RCT). Despite being the simplest possible causal setting\r\nand a perfect fit for deep learning, we theoretically find that many common choices in the\r\nliterature may lead to biased estimates. To test the practical impact of these considerations,\r\nwe recorded ISTAnt, the first real-world benchmark for causal inference downstream tasks\r\non high-dimensional observations as an RCT studying how garden ants (Lasius neglectus)\r\nrespond to microparticles applied onto their colony members by hygienic grooming.\r\nComparing 6 480 models fine-tuned from state-of-the-art visual backbones, we find that\r\nthe sampling and modeling choices significantly affect the accuracy of the causal estimate,\r\nand that classification accuracy is not a proxy thereof. We further validated the analysis,\r\nrepeating it on a synthetically generated visual data set controlling the causal model. Our\r\nresults suggest that future benchmarks should carefully consider real downstream scientific\r\nquestions, especially causal ones. Further, we highlight guidelines for representation\r\nlearning methods to help answer causal questions in the sciences.\r\nCode: https://github.com/CausalLearningAI/ISTAnt\r\nData: https://doi.org/10.6084/m9.figshare.26484934.v2"}],"scopus_import":"1","main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2405.17151"}],"publication":"38th Conference on Neural Information Processing Systems","date_created":"2025-04-06T22:01:32Z","year":"2024","citation":{"mla":"Cadei, Riccardo, et al. “Smoke and Mirrors in Causal Downstream Tasks.” <i>38th Conference on Neural Information Processing Systems</i>, vol. 37, Neural Information Processing Systems Foundation, 2024, doi:<a href=\"https://doi.org/10.52202/079017-0820\">10.52202/079017-0820</a>.","short":"R. Cadei, L. Lindorfer, S. Cremer, C. Schmid, F. Locatello, in:, 38th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2024.","ieee":"R. Cadei, L. Lindorfer, S. Cremer, C. Schmid, and F. Locatello, “Smoke and mirrors in causal downstream tasks,” in <i>38th Conference on Neural Information Processing Systems</i>, Vancouver, Canada, 2024, vol. 37.","chicago":"Cadei, Riccardo, Lukas Lindorfer, Sylvia Cremer, Cordelia Schmid, and Francesco Locatello. “Smoke and Mirrors in Causal Downstream Tasks.” In <i>38th Conference on Neural Information Processing Systems</i>, Vol. 37. Neural Information Processing Systems Foundation, 2024. <a href=\"https://doi.org/10.52202/079017-0820\">https://doi.org/10.52202/079017-0820</a>.","apa":"Cadei, R., Lindorfer, L., Cremer, S., Schmid, C., &#38; Locatello, F. (2024). Smoke and mirrors in causal downstream tasks. In <i>38th Conference on Neural Information Processing Systems</i> (Vol. 37). Vancouver, Canada: Neural Information Processing Systems Foundation. <a href=\"https://doi.org/10.52202/079017-0820\">https://doi.org/10.52202/079017-0820</a>","ista":"Cadei R, Lindorfer L, Cremer S, Schmid C, Locatello F. 2024. Smoke and mirrors in causal downstream tasks. 38th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 37.","ama":"Cadei R, Lindorfer L, Cremer S, Schmid C, Locatello F. Smoke and mirrors in causal downstream tasks. In: <i>38th Conference on Neural Information Processing Systems</i>. Vol 37. Neural Information Processing Systems Foundation; 2024. doi:<a href=\"https://doi.org/10.52202/079017-0820\">10.52202/079017-0820</a>"},"date_updated":"2026-10-21T06:45:33Z","author":[{"last_name":"Cadei","id":"0fa8b76f-72f0-11ef-b75a-a5da96e5ad6b","first_name":"Riccardo","full_name":"Cadei, Riccardo"},{"last_name":"Lindorfer","id":"85f0e6d3-06b3-11ec-8982-8c5049fa4455","first_name":"Lukas","full_name":"Lindorfer, Lukas"},{"orcid":"0000-0002-2193-3868","full_name":"Cremer, Sylvia","last_name":"Cremer","id":"2F64EC8C-F248-11E8-B48F-1D18A9856A87","first_name":"Sylvia"},{"last_name":"Schmid","first_name":"Cordelia","full_name":"Schmid, Cordelia"},{"last_name":"Locatello","first_name":"Francesco","id":"26cfd52f-2483-11ee-8040-88983bcc06d4","orcid":"0000-0002-4850-0683","full_name":"Locatello, Francesco"}],"OA_place":"repository","supplementarymaterial":"yes","dataavailabilitystatement":"Code: https://github.com/CausalLearningAI/ISTAnt\r\nData: https://doi.org/10.6084/m9.figshare.26484934.v2","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","language":[{"iso":"eng"}],"article_processing_charge":"No","oa":1,"related_material":{"link":[{"relation":"software","url":"https://github.com/CausalLearningAI/ISTAnt"}],"record":[{"relation":"earlier_version","status":"public","id":"18847"}]},"fulldoi":"https://doi.org/10.52202/079017-0820","arxiv":1,"title":"Smoke and mirrors in causal downstream tasks","publication_status":"published","OA_type":"green","type":"conference","status":"public"}]