{"id":893,"date":"2026-06-03T19:15:06","date_gmt":"2026-06-03T19:15:06","guid":{"rendered":"https:\/\/sites.nd.edu\/aim3d\/?page_id=893"},"modified":"2026-06-18T20:27:56","modified_gmt":"2026-06-18T20:27:56","slug":"may-20-2026","status":"publish","type":"page","link":"https:\/\/sites.nd.edu\/aim3d\/news-events\/may-20-2026\/","title":{"rendered":"May 20, 2026"},"content":{"rendered":"\n
\"\"<\/span>
\n
\"\"<\/a><\/figure>\n<\/div><\/div><\/div>\n\n\n\n

The AIM-3D<\/strong><\/a> team continues to advance the frontier of AI-driven causal discovery and dynamic inference. Valentina Kuskova<\/strong><\/a> presented new research at the 39th International Florida Artificial Intelligence Research Society Conference (FLAIRS-39<\/a><\/em>), introducing a novel framework for distinguishing genuinely important causal relationships from those that merely appear influential in complex nonlinear AI models.<\/p>\n\n\n\n

The presentation, titled “Beyond Coefficients: Forecast-Necessity Testing for Interpretable Causal Discovery in Nonlinear Time-Series Models<\/em><\/strong><\/a>,” addresses a growing challenge in artificial intelligence: determining whether a relationship identified by a neural network is truly necessary for prediction or simply receives a high importance score due to statistical artifacts such as persistence or redundancy.<\/p>\n\n\n\n

Developed by Valentina Kuskova<\/strong><\/a>, Dmitry Zaytsev<\/a><\/strong>, and Michael Coppedge<\/strong><\/a>, the research introduces a new formulation of the concept called forecast necessity<\/em>, which evaluates causal relevance by measuring how much predictive performance deteriorates when a specific relationship is removed from a trained model. Rather than relying solely on model coefficients or importance scores, the framework uses edge ablation combined with Diebold-Mariano statistical testing to determine whether a relationship is indispensable for forecasting.<\/p>\n\n\n\n

The study applies the methodology to a panel of 139 countries using democracy indicators from the V-Dem dataset spanning 1990\u20132024. Results demonstrate that relationships with larger neural causal scores are not necessarily the most important for forecasting. In one example, a predictor with a smaller causal score proved essential for prediction, while a higher-scoring predictor was found to be largely redundant.<\/p>\n\n\n\n

According to the researchers, the findings highlight the importance of shifting from parameter-based interpretations toward behavior-based evaluations in nonlinear machine-learning systems. The work has implications for AI interpretability, causal discovery, social science forecasting, and other domains where decisions may depend on understanding the drivers of complex dynamic systems.<\/p>\n\n\n\n

The presentation is part of the broader AIM-3D<\/strong><\/a> research program at the University of Notre Dame, which applies advanced artificial intelligence methods to study democratic development, political stability, and institutional change. The project’s recent publication pipeline<\/a><\/strong> includes forthcoming papers at top computer science conferences, KDD 2026 and ICML 2026, reflecting growing interest in combining causal inference and machine learning for the analysis of complex social systems.<\/p>\n\n\n\n

<\/p>\n","protected":false},"excerpt":{"rendered":"

The AIM-3D team continues to advance the frontier of AI-driven causal discovery and dynamic inference. Valentina Kuskova presented new research at the 39th International Florida Artificial Intelligence Research Society Conference (FLAIRS-39), introducing a novel framework for distinguishing genuinely important causal relationships from those that merely appear influential in complex nonlinear AI models. The presentation, titled […]<\/p>\n","protected":false},"author":5174,"featured_media":0,"parent":80,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-893","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/sites.nd.edu\/aim3d\/wp-json\/wp\/v2\/pages\/893","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/sites.nd.edu\/aim3d\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/sites.nd.edu\/aim3d\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/sites.nd.edu\/aim3d\/wp-json\/wp\/v2\/users\/5174"}],"replies":[{"embeddable":true,"href":"https:\/\/sites.nd.edu\/aim3d\/wp-json\/wp\/v2\/comments?post=893"}],"version-history":[{"count":5,"href":"https:\/\/sites.nd.edu\/aim3d\/wp-json\/wp\/v2\/pages\/893\/revisions"}],"predecessor-version":[{"id":962,"href":"https:\/\/sites.nd.edu\/aim3d\/wp-json\/wp\/v2\/pages\/893\/revisions\/962"}],"up":[{"embeddable":true,"href":"https:\/\/sites.nd.edu\/aim3d\/wp-json\/wp\/v2\/pages\/80"}],"wp:attachment":[{"href":"https:\/\/sites.nd.edu\/aim3d\/wp-json\/wp\/v2\/media?parent=893"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}