Using Visual Analytics to Interpret Predictive Machine Learning Models

Abstract

It is commonly believed that increasing the interpretability of a machine learning model may decrease its predictive power. However, inspecting input-output relationships of those models using visual analytics, while treating them as black-box, can help to understand the reasoning behind outcomes without sacrificing predictive quality. We identify a space of possible solutions and provide two examples of where such techniques have been successfully used in practice.

presented at 2016 ICML Workshop on Human Interpretability in Machine Learning (WHI 2016), New York, NY

References in corpus (3)

• Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
• Understanding Neural Networks Through Deep Visualization
• Visualizing and Understanding Convolutional Networks

Cited by in corpus (8)

• The Zwicky Transient Facility: Science Objectives
• Explainability Fact Sheets: A Framework for Systematic Assessment of Explainable Approaches
• The State of the Art in Enhancing Trust in Machine Learning Models with the Use of Visualizations
• Explainable Artificial Intelligence: a Systematic Review
• Inverse Classification for Comparison-based Interpretability in Machine Learning
• VINE: Visualizing Statistical Interactions in Black Box Models
• Fibres of Failure: Classifying errors in predictive processes
• i-Algebra: Towards Interactive Interpretability of Deep Neural Networks