Human-In-The-Loop Learning of Qualitative Preference Models

TL;DR

This paper introduces an interactive human-in-the-loop framework for learning and explaining qualitative preference models in combinatorial decision-making, emphasizing visualization and iterative refinement.

Contribution

It presents a novel framework combining user feedback, explainable visual models, and iterative learning for qualitative preferences.

Findings

Developed visualization tools for preference models

Enabled iterative refinement based on user feedback

Applied framework to lexicographic preference models

Abstract

In this work, we present a novel human-in-the-loop framework to help the human user understand the decision making process that involves choosing preferred options. We focus on qualitative preference models over alternatives from combinatorial domains. This framework is interactive: the user provides her behavioral data to the framework, and the framework explains the learned model to the user. It is iterative: the framework collects feedback on the learned model from the user and tries to improve it accordingly till the user terminates the iteration. In order to communicate the learned preference model to the user, we develop visualization of intuitive and explainable graphic models, such as lexicographic preference trees and forests, and conditional preference networks. To this end, we discuss key aspects of our framework for lexicographic preference models.