Guardado en:
Detalles Bibliográficos
Autores principales: Gavito, Andrea Treviño, Klabjan, Diego, Utke, Jean
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:https://arxiv.org/abs/2302.14278
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911901087170560
author Gavito, Andrea Treviño
Klabjan, Diego
Utke, Jean
author_facet Gavito, Andrea Treviño
Klabjan, Diego
Utke, Jean
contents We propose a graph-oriented attention-based explainability method for tabular data. Tasks involving tabular data have been solved mostly using traditional tree-based machine learning models which have the challenges of feature selection and engineering. With that in mind, we consider a transformer architecture for tabular data, which is amenable to explainability, and present a novel way to leverage self-attention mechanism to provide explanations by taking into account the attention matrices of all heads and layers as a whole. The matrices are mapped to a graph structure where groups of features correspond to nodes and attention values to arcs. By finding the maximum probability paths in the graph, we identify groups of features providing larger contributions to explain the model's predictions. To assess the quality of multi-layer attention-based explanations, we compare them with popular attention-, gradient-, and perturbation-based explanability methods.
format Preprint
id arxiv_https___arxiv_org_abs_2302_14278
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Multi-Layer Attention-Based Explainability via Transformers for Tabular Data
Gavito, Andrea Treviño
Klabjan, Diego
Utke, Jean
Machine Learning
Artificial Intelligence
We propose a graph-oriented attention-based explainability method for tabular data. Tasks involving tabular data have been solved mostly using traditional tree-based machine learning models which have the challenges of feature selection and engineering. With that in mind, we consider a transformer architecture for tabular data, which is amenable to explainability, and present a novel way to leverage self-attention mechanism to provide explanations by taking into account the attention matrices of all heads and layers as a whole. The matrices are mapped to a graph structure where groups of features correspond to nodes and attention values to arcs. By finding the maximum probability paths in the graph, we identify groups of features providing larger contributions to explain the model's predictions. To assess the quality of multi-layer attention-based explanations, we compare them with popular attention-, gradient-, and perturbation-based explanability methods.
title Multi-Layer Attention-Based Explainability via Transformers for Tabular Data
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2302.14278