Coactive Critiquing: Elicitation of Preferences and Features

TL;DR

This paper introduces Coactive Critiquing, a preference elicitation method that combines user feedback and critiques to improve learning in complex, dynamic choice scenarios, supporting constructive tasks with on-the-fly catalog generation.

Contribution

It extends Coactive Learning by incorporating user critiques and dynamically expanding feature space, enabling more effective preference elicitation in complex, constructive environments.

Findings

Lowered average regret in preference learning tasks

Effective integration of critiques into feature space

Promising empirical results demonstrating approach potential

Abstract

When faced with complex choices, users refine their own preference criteria as they explore the catalogue of options. In this paper we propose an approach to preference elicitation suited for this scenario. We extend Coactive Learning, which iteratively collects manipulative feedback, to optionally query example critiques. User critiques are integrated into the learning model by dynamically extending the feature space. Our formulation natively supports constructive learning tasks, where the option catalogue is generated on-the-fly. We present an upper bound on the average regret suffered by the learner. Our empirical analysis highlights the promise of our approach.