To Ask or Not to Ask: Learning to Require Human Feedback

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Pugnana, Andrea, De Toni, Giovanni, Barbera, Cesare, Pellungrini, Roberto, Lepri, Bruno, Passerini, Andrea
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911200254623744
author Pugnana, Andrea
De Toni, Giovanni
Barbera, Cesare
Pellungrini, Roberto
Lepri, Bruno
Passerini, Andrea
author_facet Pugnana, Andrea
De Toni, Giovanni
Barbera, Cesare
Pellungrini, Roberto
Lepri, Bruno
Passerini, Andrea
contents Developing decision-support systems that complement human performance in classification tasks remains an open challenge. A popular approach, Learning to Defer (LtD), allows a Machine Learning (ML) model to pass difficult cases to a human expert. However, LtD treats humans and ML models as mutually exclusive decision-makers, restricting the expert contribution to mere predictions. To address this limitation, we propose Learning to Ask (LtA), a new framework that handles both when and how to incorporate expert input in an ML model. LtA is based on a two-part architecture: a standard ML model and an enriched model trained with additional expert human feedback, with a formally optimal strategy for selecting when to query the enriched model. We provide two practical implementations of LtA: a sequential approach, which trains the models in stages, and a joint approach, which optimises them simultaneously. For the latter, we design surrogate losses with realisable-consistency guarantees. Our experiments with synthetic and real expert data demonstrate that LtA provides a more flexible and powerful foundation for effective human-AI collaboration.
format Preprint
id arxiv_https___arxiv_org_abs_2510_08314
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle To Ask or Not to Ask: Learning to Require Human Feedback
Pugnana, Andrea
De Toni, Giovanni
Barbera, Cesare
Pellungrini, Roberto
Lepri, Bruno
Passerini, Andrea
Machine Learning
Human-Computer Interaction
Developing decision-support systems that complement human performance in classification tasks remains an open challenge. A popular approach, Learning to Defer (LtD), allows a Machine Learning (ML) model to pass difficult cases to a human expert. However, LtD treats humans and ML models as mutually exclusive decision-makers, restricting the expert contribution to mere predictions. To address this limitation, we propose Learning to Ask (LtA), a new framework that handles both when and how to incorporate expert input in an ML model. LtA is based on a two-part architecture: a standard ML model and an enriched model trained with additional expert human feedback, with a formally optimal strategy for selecting when to query the enriched model. We provide two practical implementations of LtA: a sequential approach, which trains the models in stages, and a joint approach, which optimises them simultaneously. For the latter, we design surrogate losses with realisable-consistency guarantees. Our experiments with synthetic and real expert data demonstrate that LtA provides a more flexible and powerful foundation for effective human-AI collaboration.
title To Ask or Not to Ask: Learning to Require Human Feedback
topic Machine Learning
Human-Computer Interaction
url https://arxiv.org/abs/2510.08314