Enforcing Consistency and Fairness in Multi-level Hierarchical Classification with a Mask-based Output Layer
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866908275900940288 |
|---|---|
| author | Chen, Shijing Jameel, Shoaib Bouadjenek, Mohamed Reda Tang, Feilong Naseem, Usman Suleiman, Basem Hacid, Hakim Salim, Flora D. Razzak, Imran |
| author_facet | Chen, Shijing Jameel, Shoaib Bouadjenek, Mohamed Reda Tang, Feilong Naseem, Usman Suleiman, Basem Hacid, Hakim Salim, Flora D. Razzak, Imran |
| contents | Traditional Multi-level Hierarchical Classification (MLHC) classifiers often rely on backbone models with $n$ independent output layers. This structure tends to overlook the hierarchical relationships between classes, leading to inconsistent predictions that violate the underlying taxonomy. Additionally, once a backbone architecture for an MLHC classifier is selected, adapting the model to accommodate new tasks can be challenging. For example, incorporating fairness to protect sensitive attributes within a hierarchical classifier necessitates complex adjustments to maintain the class hierarchy while enforcing fairness constraints. In this paper, we extend this concept to hierarchical classification by introducing a fair, model-agnostic layer designed to enforce taxonomy and optimize specific objectives, including consistency, fairness, and exact match. Our evaluations demonstrate that the proposed layer not only improves the fairness of predictions but also enforces the taxonomy, resulting in consistent predictions and superior performance. Compared to Large Language Models (LLMs) employing in-processing de-biasing techniques and models without any bias correction, our approach achieves better outcomes in both fairness and accuracy, making it particularly valuable in sectors like e-commerce, healthcare, and education, where predictive reliability is crucial. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_15566 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Enforcing Consistency and Fairness in Multi-level Hierarchical Classification with a Mask-based Output Layer Chen, Shijing Jameel, Shoaib Bouadjenek, Mohamed Reda Tang, Feilong Naseem, Usman Suleiman, Basem Hacid, Hakim Salim, Flora D. Razzak, Imran Machine Learning Traditional Multi-level Hierarchical Classification (MLHC) classifiers often rely on backbone models with $n$ independent output layers. This structure tends to overlook the hierarchical relationships between classes, leading to inconsistent predictions that violate the underlying taxonomy. Additionally, once a backbone architecture for an MLHC classifier is selected, adapting the model to accommodate new tasks can be challenging. For example, incorporating fairness to protect sensitive attributes within a hierarchical classifier necessitates complex adjustments to maintain the class hierarchy while enforcing fairness constraints. In this paper, we extend this concept to hierarchical classification by introducing a fair, model-agnostic layer designed to enforce taxonomy and optimize specific objectives, including consistency, fairness, and exact match. Our evaluations demonstrate that the proposed layer not only improves the fairness of predictions but also enforces the taxonomy, resulting in consistent predictions and superior performance. Compared to Large Language Models (LLMs) employing in-processing de-biasing techniques and models without any bias correction, our approach achieves better outcomes in both fairness and accuracy, making it particularly valuable in sectors like e-commerce, healthcare, and education, where predictive reliability is crucial. |
| title | Enforcing Consistency and Fairness in Multi-level Hierarchical Classification with a Mask-based Output Layer |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2503.15566 |