{"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/timeshap-explaining-recurrent-models-through","title":"TimeSHAP: Explaining Recurrent Models through Sequence Perturbations","arxiv_id":"2012.00073","date":"2020-11-30","proceeding":null,"authors":["João Bento","Pedro Saleiro","André F. Cruz","Mário A. T. Figueiredo","Pedro Bizarro"],"abstract":"Although recurrent neural networks (RNNs) are state-of-the-art in numerous sequential decision-making tasks, there has been little research on explaining their predictions. In this work, we present TimeSHAP, a model-agnostic recurrent explainer that builds upon KernelSHAP and extends it to the sequential domain. TimeSHAP computes feature-, timestep-, and cell-level attributions. As sequences may be arbitrarily long, we further propose a pruning method that is shown to dramatically decrease both its computational cost and the variance of its attributions. We use TimeSHAP to explain the predictions of a real-world bank account takeover fraud detection RNN model, and draw key insights from its explanations: i) the model identifies important features and events aligned with what fraud analysts consider cues for account takeover; ii) positive predicted sequences can be pruned to only 10% of the original length, as older events have residual attribution values; iii) the most recent input event of positive predictions only contributes on average to 41% of the model's score; iv) notably high attribution to client's age, suggesting a potential discriminatory reasoning, later confirmed as higher false positive rates for older clients.","url_abs":"https://arxiv.org/abs/2012.00073v2","url_pdf":"https://arxiv.org/pdf/2012.00073v2.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":"timeshap-explaining-recurrent-models-through","repo_url":"https://github.com/feedzai/timeshap","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"decision-making","task_name":"Decision Making"},{"task_slug":"feature-importance","task_name":"Feature Importance"},{"task_slug":"fraud-detection","task_name":"Fraud Detection"},{"task_slug":"sequential-decision-making","task_name":"Sequential Decision Making"},{"task_slug":"time-series","task_name":"Time Series Analysis"},{"task_slug":"time-series-classification","task_name":"Time Series Classification"}],"methods":[{"method_slug":"pruning","method_name":"Pruning"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2012.00073","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}