Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2506.06455 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866918470347653120 |
|---|---|
| author | Banegas-Luna, Antonio Jesús Pérez-Sánchez, Horacio Martínez-Cortés, Carlos |
| author_facet | Banegas-Luna, Antonio Jesús Pérez-Sánchez, Horacio Martínez-Cortés, Carlos |
| contents | While predictive accuracy is often prioritized in machine learning (ML) models, interpretability remains essential in scientific and high-stakes domains. However, diverse interpretability algorithms frequently yield conflicting explanations, highlighting the need for consensus to harmonize results. In this study, six ML models were trained on six synthetic datasets with known ground truths, utilizing various model-agnostic interpretability techniques. Consensus explanations were generated using established methods and a novel approach: WISCA (Weighted Scaled Consensus Attributions), which integrates class probability and normalized attributions. WISCA consistently aligned with the most reliable individual method, underscoring the value of robust consensus strategies in improving explanation reliability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_06455 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | WISCA: A Consensus-Based Approach to Harmonizing Interpretability in Tabular Datasets Banegas-Luna, Antonio Jesús Pérez-Sánchez, Horacio Martínez-Cortés, Carlos Machine Learning Artificial Intelligence While predictive accuracy is often prioritized in machine learning (ML) models, interpretability remains essential in scientific and high-stakes domains. However, diverse interpretability algorithms frequently yield conflicting explanations, highlighting the need for consensus to harmonize results. In this study, six ML models were trained on six synthetic datasets with known ground truths, utilizing various model-agnostic interpretability techniques. Consensus explanations were generated using established methods and a novel approach: WISCA (Weighted Scaled Consensus Attributions), which integrates class probability and normalized attributions. WISCA consistently aligned with the most reliable individual method, underscoring the value of robust consensus strategies in improving explanation reliability. |
| title | WISCA: A Consensus-Based Approach to Harmonizing Interpretability in Tabular Datasets |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2506.06455 |