View a PDF of the paper titled Statistical Deficiency for Task Inclusion Estimation, by Lo\”ic Fosse and Fr\’ed\’eric B\’echet and Beno\^it Favre and G\’eraldine Damnati and Gw\’enol\’e Lecorv\’e and Maxime Darrin and Philippe Formont and Pablo Piantanida
Abstract:Tasks are central in machine learning, as they are the most natural objects to assess the capabilities of current models. The trend is to build general models able to address any task. Even though transfer learning and multitask learning try to leverage the underlying task space, no well-founded tools are available to study its structure. This study proposes a theoretically grounded setup to define the notion of task and to compute the {\bf inclusion} between two tasks from a statistical deficiency point of view. We propose a tractable proxy as information sufficiency to estimate the degree of inclusion between tasks, show its soundness on synthetic data, and use it to reconstruct empirically the classic NLP pipeline.
Submission history
From: Loïc Fosse [view email]
[v1]
Fri, 7 Mar 2025 15:00:28 UTC (5,370 KB)
[v2]
Thu, 13 Mar 2025 08:41:29 UTC (5,337 KB)
