Extracting PAC Decision Trees from Black Box Binary Classifiers: The Gender Bias Case Study on BERT-based Language Models

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Ozaki, Ana, Confalonieri, Roberto, Guimarães, Ricardo, Imenes, Anders
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916991858638848
author Ozaki, Ana
Confalonieri, Roberto
Guimarães, Ricardo
Imenes, Anders
author_facet Ozaki, Ana
Confalonieri, Roberto
Guimarães, Ricardo
Imenes, Anders
contents Decision trees are a popular machine learning method, known for their inherent explainability. In Explainable AI, decision trees can be used as surrogate models for complex black box AI models or as approximations of parts of such models. A key challenge of this approach is determining how accurately the extracted decision tree represents the original model and to what extent it can be trusted as an approximation of their behavior. In this work, we investigate the use of the Probably Approximately Correct (PAC) framework to provide a theoretical guarantee of fidelity for decision trees extracted from AI models. Based on theoretical results from the PAC framework, we adapt a decision tree algorithm to ensure a PAC guarantee under certain conditions. We focus on binary classification and conduct experiments where we extract decision trees from BERT-based language models with PAC guarantees. Our results indicate occupational gender bias in these models.
format Preprint
id arxiv_https___arxiv_org_abs_2412_10513
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Extracting PAC Decision Trees from Black Box Binary Classifiers: The Gender Bias Case Study on BERT-based Language Models
Ozaki, Ana
Confalonieri, Roberto
Guimarães, Ricardo
Imenes, Anders
Artificial Intelligence
Machine Learning
Decision trees are a popular machine learning method, known for their inherent explainability. In Explainable AI, decision trees can be used as surrogate models for complex black box AI models or as approximations of parts of such models. A key challenge of this approach is determining how accurately the extracted decision tree represents the original model and to what extent it can be trusted as an approximation of their behavior. In this work, we investigate the use of the Probably Approximately Correct (PAC) framework to provide a theoretical guarantee of fidelity for decision trees extracted from AI models. Based on theoretical results from the PAC framework, we adapt a decision tree algorithm to ensure a PAC guarantee under certain conditions. We focus on binary classification and conduct experiments where we extract decision trees from BERT-based language models with PAC guarantees. Our results indicate occupational gender bias in these models.
title Extracting PAC Decision Trees from Black Box Binary Classifiers: The Gender Bias Case Study on BERT-based Language Models
topic Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2412.10513