Explaining Neural Networks in Preference Learning: a Post-hoc Inductive Logic Programming Approach

TL;DR

This paper introduces a method to interpret neural networks in preference learning by approximating them with inductive logic programming, using PCA for dimensionality reduction to improve explanation transparency.

Contribution

It presents a novel approach combining ILASP and PCA to approximate neural networks in preference learning, enhancing interpretability and computational efficiency.

Findings

ILASP can effectively approximate neural networks in preference learning.

Dimensionality reduction with PCA improves explanation clarity and computational performance.

The approach balances fidelity to the original model with interpretability.

Abstract

In this paper, we propose using Learning from Answer Sets to approximate black-box models, such as Neural Networks (NN), in the specific case of learning user preferences. We specifically explore the use of ILASP (Inductive Learning of Answer Set Programs) to approximate preference learning systems through weak constraints. We have created a dataset on user preferences over a set of recipes, which is used to train the NNs that we aim to approximate with ILASP. Our experiments investigate ILASP both as a global and a local approximator of the NNs. These experiments address the challenge of approximating NNs working on increasingly high-dimensional feature spaces while achieving appropriate fidelity on the target model and limiting the increase in computational time. To handle this challenge, we propose a preprocessing step that exploits Principal Component Analysis to reduce the dataset's dimensionality while keeping our explanations transparent. Under consideration for publication in Theory and Practice of Logic Programming (TPLP).