An Explainable Multi-Task Similarity Measure: Integrating Accumulated Local Effects and Weighted Fréchet Distance Journal: Knowledge-Based Systems
Pablo Hidalgo, Daniel Rodriguez · 2026
Read An Explainable Multi-Task Similarity Measure: Integrating Accumulated Local Effects and Weighted Fréchet Distance Journal: Knowledge-Based Systems by Pablo Hidalgo, Daniel Rodriguez (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-07966.txt.
Description
Multi-task learning promises knowledge transfer between tasks, but deciding which tasks are similar, and why, remains an open question. Pablo Hidalgo and Daniel Rodriguez propose a similarity measure built from explainable AI: accumulated local effects curves, which show how each feature influences predictions, compared with a distribution-weighted Fréchet distance.
The measure incorporates feature importance, works whether tasks are trained separately or together, and applies across model types, with a scaling factor for differing predictive performance. The authors also offer recommendations for using it in practice. Researchers grouping tasks for transfer learning gain an interpretable, model-agnostic way to justify which tasks should learn together.
Catalog record
| CREATOR | Pablo Hidalgo, Daniel Rodriguez |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 87 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-07966