Disentangling Neural Disjunctive Normal Form Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Baugh, Kexin Gu, Perreault, Vincent, Baugh, Matthew, Dickens, Luke, Inoue, Katsumi, Russo, Alessandra
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909715116589056
author Baugh, Kexin Gu
Perreault, Vincent
Baugh, Matthew
Dickens, Luke
Inoue, Katsumi
Russo, Alessandra
author_facet Baugh, Kexin Gu
Perreault, Vincent
Baugh, Matthew
Dickens, Luke
Inoue, Katsumi
Russo, Alessandra
contents Neural Disjunctive Normal Form (DNF) based models are powerful and interpretable approaches to neuro-symbolic learning and have shown promising results in classification and reinforcement learning settings without prior knowledge of the tasks. However, their performance is degraded by the thresholding of the post-training symbolic translation process. We show here that part of the performance degradation during translation is due to its failure to disentangle the learned knowledge represented in the form of the networks' weights. We address this issue by proposing a new disentanglement method; by splitting nodes that encode nested rules into smaller independent nodes, we are able to better preserve the models' performance. Through experiments on binary, multiclass, and multilabel classification tasks (including those requiring predicate invention), we demonstrate that our disentanglement method provides compact and interpretable logical representations for the neural DNF-based models, with performance closer to that of their pre-translation counterparts. Our code is available at https://github.com/kittykg/disentangling-ndnf-classification.
format Preprint
id arxiv_https___arxiv_org_abs_2507_10546
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Disentangling Neural Disjunctive Normal Form Models
Baugh, Kexin Gu
Perreault, Vincent
Baugh, Matthew
Dickens, Luke
Inoue, Katsumi
Russo, Alessandra
Machine Learning
Artificial Intelligence
Neural Disjunctive Normal Form (DNF) based models are powerful and interpretable approaches to neuro-symbolic learning and have shown promising results in classification and reinforcement learning settings without prior knowledge of the tasks. However, their performance is degraded by the thresholding of the post-training symbolic translation process. We show here that part of the performance degradation during translation is due to its failure to disentangle the learned knowledge represented in the form of the networks' weights. We address this issue by proposing a new disentanglement method; by splitting nodes that encode nested rules into smaller independent nodes, we are able to better preserve the models' performance. Through experiments on binary, multiclass, and multilabel classification tasks (including those requiring predicate invention), we demonstrate that our disentanglement method provides compact and interpretable logical representations for the neural DNF-based models, with performance closer to that of their pre-translation counterparts. Our code is available at https://github.com/kittykg/disentangling-ndnf-classification.
title Disentangling Neural Disjunctive Normal Form Models
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2507.10546