How to Open a Black Box Classifier for Tabular Data

被引:6
|
作者
Walters, Bradley [1 ]
Ortega-Martorell, Sandra [1 ]
Olier, Ivan [1 ]
Lisboa, Paulo J. G. [1 ]
机构
[1] Liverpool John Moores Univ, Sch Comp Sci & Math, Liverpool L3 2AF, England
关键词
ANOVA; Shapley values; self-explaining neural networks; generalised additive models; interpretability;
D O I
10.3390/a16040181
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
A lack of transparency in machine learning models can limit their application. We show that analysis of variance (ANOVA) methods extract interpretable predictive models from them. This is possible because ANOVA decompositions represent multivariate functions as sums of functions of fewer variables. Retaining the terms in the ANOVA summation involving functions of only one or two variables provides an efficient method to open black box classifiers. The proposed method builds generalised additive models (GAMs) by application of L1 regularised logistic regression to the component terms retained from the ANOVA decomposition of the logit function. The resulting GAMs are derived using two alternative measures, Dirac and Lebesgue. Both measures produce functions that are smooth and consistent. The term partial responses in structured models (PRiSM) describes the family of models that are derived from black box classifiers by application of ANOVA decompositions. We demonstrate their interpretability and performance for the multilayer perceptron, support vector machines and gradient-boosting machines applied to synthetic data and several real-world data sets, namely Pima Diabetes, German Credit Card, and Statlog Shuttle from the UCI repository. The GAMs are shown to be compliant with the basic principles of a formal framework for interpretability.
引用
收藏
页数:26
相关论文
共 50 条
  • [1] Open your black box classifier
    Lisboa, Paulo Jorge Gomes
    HEALTHCARE TECHNOLOGY LETTERS, 2024, 11 (04) : 210 - 212
  • [2] AuthSHAP - Authentication Vulnerability Detection on Tabular Data in Black Box Setting
    Das, Debasmita
    Kaytal, Yatin
    Ganesh, Ram, V
    Bhattacharya, Rohit
    Ranjan, Rajesh Kumar
    ICAIF 2021: THE SECOND ACM INTERNATIONAL CONFERENCE ON AI IN FINANCE, 2021,
  • [3] A FEW INSTRUCTIONS ON HOW TO OPEN THE BLACK BOX OF FEMINIST KNOWLEDGE
    Hernandez Garcia, Jone M.
    DISPARIDADES-REVISTA DE ANTROPOLOGIA, 2019, 74 (01):
  • [4] A Logic of "Black Box" Classifier Systems
    Liu, Xinghan
    Lorini, Emiliano
    LOGIC, LANGUAGE, INFORMATION, AND COMPUTATION, WOLLIC 2022, 2022, 13468 : 158 - 174
  • [5] Microalgae protein digestibility: How to crack open the black box?
    van de Walle, Simon
    Broucke, Keshia
    Baune, Marie-Christin
    Terjung, Nino
    Van Royen, Geert
    Boukid, Fatma
    CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION, 2024, 64 (20) : 7149 - 7171
  • [6] Tiny classifier circuits as accelerators for classification of tabular data
    Iordanou, Konstantinos
    Lujan, Mikel
    NATURE ELECTRONICS, 2024, 7 (05): : 334 - 335
  • [7] Evading Open Science: The Black Box of Student Data Collection
    Ludwig, Tobias
    Altenmueller, Marlene Sophie
    Schramm, Leonhard Falk Florentin
    Twardawski, Mathias
    SOCIAL PSYCHOLOGICAL BULLETIN, 2023, 18
  • [8] Fee-for-data: A strategy to open the HMO black box
    Welch, WP
    Welch, HG
    HEALTH AFFAIRS, 1995, 14 (04) : 104 - 116
  • [9] ChatGPT is a black box: how AI research can break it open
    不详
    NATURE, 2023, 619 (7971) : 671 - 672