One Explanation Does Not Fit All: A Toolkit and Taxonomy of AI Explainability Techniques

2019/09/06 by Vijay Arya, Arya, Vijay, Rachel K. E. Bellamy +40 · 24 citations
Computer Science · Decision Sciences · Mathematics · #Adversarial Robustness in Machine Learning #Artificial Intelligence (cs.AI) #Computer Vision and Pattern Recognition (cs.CV) #Explainable Artificial Intelligence (XAI) #FOS: Computer and information sciences #Human-Computer Interaction (cs.HC) #Machine Learning (stat.ML) #Scientific Computing and Data Management #cs.AI #cs.CV #cs.HC #stat.ML

paper · pdf · doi:10.48550/arxiv.1909.03012

openalex publication_date 2019/09/06 · arxiv created 2019/09/14 · arxiv updated 2019/09/17 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

As artificial intelligence and machine learning algorithms make further inroads into society, calls are increasing from multiple stakeholders for these algorithms to explain their outputs. At the same time, these stakeholders, whether they be affected citizens, government regulators, domain experts, or system developers, present different requirements for explanations. Toward addressing these needs, we introduce AI Explainability 360 (http://aix360.mybluemix.net/), an open-source software toolkit featuring eight diverse and state-of-the-art explainability methods and two evaluation metrics. Equally important, we provide a taxonomy to help entities requiring explanations to navigate the space of explanation methods, not only those in the toolkit but also in the broader literature on explainability. For data scientists and other users of the toolkit, we have implemented an extensible software architecture that organizes methods according to their place in the AI modeling pipeline. We also discuss enhancements to bring research innovations closer to consumers of explanations, ranging from simplified, more accessible versions of algorithms, to tutorials and an interactive web demo to introduce AI explainability to different audiences and application domains. Together, our toolkit and taxonomy can help identify gaps where more explainability methods are needed and provide a platform to incorporate them as they are developed.

Citations

• Distilling the Knowledge in a Neural Network
• Boolean Decision Rules via Column Generation
• Towards Robust Interpretability with Self-Explaining Neural Networks
• Improving Simple Models with Confidence Profiles
• A Survey Of Methods For Explaining Black Box Models
• Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps

Cited by

• T-FIX: Text-Based Explanations with Features Interpretable to eXperts
• "It feels like hard work trying to talk to it": Understanding Older Adults' Experiences of Encountering and Repairing Conversational Breakdowns with AI Systems
• Onto-Epistemological Analysis of AI Explanations
• AttentionDep: Domain-Aware Attention for Explainable Depression Severity Assessment
• Outlier Detection in Plantar Pressure: Human-Centered Comparison of Statistical Parametric Mapping and Explainable Machine Learning
• Holistic Explainable AI (H-XAI): Extending Transparency Beyond Developers in AI-Driven Decision Making
• MetaExplainer: A Framework to Generate Multi-Type User-Centered Explanations for AI Systems
• Your Model Is Unfair, Are You Even Aware? Inverse Relationship Between Comprehension and Trust in Explainability Visualizations of Biased ML Models
• Distilling Knowledge from Large Language Models: A Concept Bottleneck Model for Hate and Counter Speech Recognition
• Explainability in machine learning: a pedagogical perspective
• Controlling Neural Networks with Rule Representations
• Human-Centered Explainability in Interactive Information Systems: A Survey
• Toward the Explainability of Protein Language Models
• The Metacognitive Demands and Opportunities of Generative AI
• Explainability in Context: A Multilevel Framework Aligning AI Explanations with Stakeholder with LLMs
• MOSS: Multi-Objective Optimization for Stable Rule Sets
• Transparency, Auditability and eXplainability of Machine Learning Models\n in Credit Scoring
• Resisting Out-of-Distribution Data Problem in Perturbation of XAI
• Towards Uncertainty Aware Task Delegation and Human-AI Collaborative Decision-Making
• A User Study Evaluating Argumentative Explanations in Diagnostic Decision Support
• Evaluating Explanation Quality in X-IDS Using Feature Alignment Metrics
• ODExAI: A Comprehensive Object Detection Explainable AI Evaluation
• AI Explainability 360: Impact and Design
• What Makes for a Good Saliency Map? Comparing Strategies for Evaluating Saliency Maps in Explainable AI (XAI)

Related

• Towards Quantification of Explainability in Explainable Artificial Intelligence Methods
• How to explain it to data scientists? A mixed-methods user study about explainable AI, using mental models for explanations
• Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AI
• A Comprehensive Perspective on Explainable AI across the Machine Learning Workflow
• A Survey of Explainable AI and Proposal for a Discipline of Explanation Engineering