Redirigiendo al acceso original de articulo en 22 segundos...
Inicio  /  Information  /  Vol: 11 Par: 3 (2020)  /  Artículo
ARTÍCULO
TITULO

A Responsible Machine Learning Workflow with Focus on Interpretable Models, Post-hoc Explanation, and Discrimination Testing

Navdeep Gill    
Patrick Hall    
Kim Montgomery and Nicholas Schmidt    

Resumen

This manuscript outlines a viable approach for training and evaluating machine learning systems for high-stakes, human-centered, or regulated applications using common Python programming tools. The accuracy and intrinsic interpretability of two types of constrained models, monotonic gradient boosting machines and explainable neural networks, a deep learning architecture well-suited for structured data, are assessed on simulated data and publicly available mortgage data. For maximum transparency and the potential generation of personalized adverse action notices, the constrained models are analyzed using post-hoc explanation techniques including plots of partial dependence and individual conditional expectation and with global and local Shapley feature importance. The constrained model predictions are also tested for disparate impact and other types of discrimination using measures with long-standing legal precedents, adverse impact ratio, marginal effect, and standardized mean difference, along with straightforward group fairness measures. By combining interpretable models, post-hoc explanations, and discrimination testing with accessible software tools, this text aims to provide a template workflow for machine learning applications that require high accuracy and interpretability and that mitigate risks of discrimination.

 Artículos similares

       
 
Luke Balcombe    
Artificial intelligence (AI) chatbots have gained prominence since 2022. Powered by big data, natural language processing (NLP) and machine learning (ML) algorithms, they offer the potential to expand capabilities, improve productivity and provide guidan... ver más
Revista: Informatics

 
Mihai-Virgil Nichita, Maria-Alexandra Paun, Vladimir-Alexandru Paun and Viorel-Puiu Paun    
This paper introduces an AI model designed for the diagnosis and monitoring of the SARS-CoV-2 virus. The present artificial intelligence (AI) model founded on the machine learning concept was created for the identification/recognition, keeping under obse... ver más
Revista: Computers

 
Mudassir M. Rashid, Mohammad Reza Askari, Canyu Chen, Yueqing Liang, Kai Shu and Ali Cinar    
Artificial intelligence (AI) algorithms can provide actionable insights for clinical decision-making and managing chronic diseases. The treatment and management of complex chronic diseases, such as diabetes, stands to benefit from novel AI algorithms ana... ver más
Revista: Algorithms

 
Matheus Alencar da Silva, Bonfim Amaro Junior, Ramon Rudá Brito Medeiros and Plácido Rogério Pinheiro    
Three-dimensional printing has advantages, such as an excellent flexibility in producing parts from the digital model, enabling the fabrication of different geometries that are both simple or complex, using low-cost materials and generating little residu... ver más
Revista: Algorithms

 
Alessandro Bonforte, Flavio Cannavò, Salvatore Gambino and Francesco Guglielmino    
We propose a multi-temporal-scale analysis of ground deformation data using both high-rate tilt and GNSS measurements and the DInSAR and daily GNSS solutions in order to investigate a sequence of four paroxysmal episodes of the Voragine crater occurring ... ver más
Revista: Applied Sciences