| Autor: |
Wang, Xiangmeng, Li, Qian, Yu, Dianer, Li, Qing, Xu, Guandong |
| Zdroj: |
ACM Transactions on Information Systems; Jul2024, Vol. 42 Issue 4, p1-30, 30p |
| Abstrakt: |
The article introduces CFairER, a Counterfactual Explanation for Fairness in recommendation systems, utilizing causal inference to generate attribute-level counterfactual explanations, addressing the challenge of explaining unfair recommendations and enhancing model trust. Topics include fairness diagnostics, off-policy reinforcement learning for generating high-quality explanations, and attentive action pruning for narrowing the search space of candidate counterfactuals. |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
