When learning becomes impossible - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

When learning becomes impossible

Résumé

We formally analyze an epistemic bias we call interpretive blindness (IB), in which under certain conditions a learner will be incapable of learning. IB is now common in our society, but it is a natural consequence of Bayesian inference and what we argue are mild assumptions about the relation between belief and evidence. IB a special problem for learning from testimony, in which one acquires information only from text or conversation. We show that IB follows from a codependence between background beliefs and interpretation in a Bayesian setting and the nature of contemporary testimony. We argue that a particular characteristic of contemporary testimony, argumentative completeness, can preclude learning in hierarchical Bayesian settings, even in the presence of constraints that are designed to promote good epistemic practices.
Fichier principal
Vignette du fichier
When_learning_becomes_impossible.pdf (511.5 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03873806 , version 1 (27-11-2022)

Identifiants

Citer

Nicholas Asher, Julie Hunter. When learning becomes impossible. 5th ACM Conference on Fairness, Accountability, and Transparency (ACM FAccT 2022), ACM: Association for Computing Machinery, Jun 2022, Séoul, South Korea. pp.107-116, ⟨10.1145/3531146.3533078⟩. ⟨hal-03873806⟩
15 Consultations
30 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More