Bound Propagation meets Constraint Simplification: Improving Logic-based XAI for Neural Networks
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917305961676800 |
|---|---|
| author | Gomes, Ronaldo Ribeiro, Jairo Queiroz, Luiz Rocha, Thiago Alves |
| author_facet | Gomes, Ronaldo Ribeiro, Jairo Queiroz, Luiz Rocha, Thiago Alves |
| contents | Logic-based methods for explaining neural network decisions offer formal guarantees of correctness and non-redundancy, but they often suffer from high computational costs, especially for large networks. In this work, we improve the efficiency of such methods by combining bound propagation with constraint simplification. These simplifications, derived from the propagation, tighten neuron bounds and eliminate unnecessary binary variables, making the explanation process more efficient. Our experiments suggest that combining these techniques reduces explanation time by up to 89.26\%, particularly for larger neural networks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_01923 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Bound Propagation meets Constraint Simplification: Improving Logic-based XAI for Neural Networks Gomes, Ronaldo Ribeiro, Jairo Queiroz, Luiz Rocha, Thiago Alves Logic in Computer Science Machine Learning Logic-based methods for explaining neural network decisions offer formal guarantees of correctness and non-redundancy, but they often suffer from high computational costs, especially for large networks. In this work, we improve the efficiency of such methods by combining bound propagation with constraint simplification. These simplifications, derived from the propagation, tighten neuron bounds and eliminate unnecessary binary variables, making the explanation process more efficient. Our experiments suggest that combining these techniques reduces explanation time by up to 89.26\%, particularly for larger neural networks. |
| title | Bound Propagation meets Constraint Simplification: Improving Logic-based XAI for Neural Networks |
| topic | Logic in Computer Science Machine Learning |
| url | https://arxiv.org/abs/2603.01923 |