Explainable Artificial Intelligence: a Systematic Review

Abstract

Explainable Artificial Intelligence (XAI) has experienced a significant growth over the last few years. This is due to the widespread application of machine learning, particularly deep learning, that has led to the development of highly accurate models but lack explainability and interpretability. A plethora of methods to tackle this problem have been proposed, developed and tested. This systematic review contributes to the body of knowledge by clustering these methods with a hierarchical classification system with four main clusters: review articles, theories and notions, methods and their evaluation. It also summarises the state-of-the-art in XAI and recommends future research directions.

arXiv comment78 pages, 18 figures, journal paper to be submitted to Information Fusion

References in corpus (40)

• Axiomatic Attribution for Deep Networks
• Methods for Interpreting and Understanding Deep Neural Networks
• Learning Important Features Through Propagating Activation Differences
• Understanding Neural Networks Through Deep Visualization
• Explaining NonLinear Classification Decisions with Deep Taylor Decomposition
• Understanding Black-box Predictions via Influence Functions
• Explainable Artificial Intelligence: Understanding, Visualizing and Interpreting Deep Learning Models
• Visualizing and Understanding Recurrent Networks
• Interpretable classifiers using rules and Bayesian analysis: Building a better stroke prediction model
• SmoothGrad: removing noise by adding noise
• Model-Agnostic Interpretability of Machine Learning
• Sanity Checks for Saliency Maps
• Towards Explainable Artificial Intelligence
• Visualizing Deep Neural Network Decisions: Prediction Difference Analysis
• Learning how to explain neural networks: PatternNet and PatternAttribution
• "What is Relevant in a Text Document?": An Interpretable Machine Learning Approach
• explAIner: A Visual Analytics Framework for Interactive and Explainable Machine Learning
• Prototype selection for interpretable classification
• SVCCA: Singular Vector Canonical Correlation Analysis for Deep Learning Dynamics and Interpretability
• Manifold: A Model-Agnostic Framework for Interpretation and Diagnosis of Machine Learning Models
• Explainable AI: Beware of Inmates Running the Asylum Or: How I Learnt to Stop Worrying and Love the Social and Behavioural Sciences
• Understanding intermediate layers using linear classifier probes
• Beyond Sparsity: Tree Regularization of Deep Models for Interpretability
• Interpretability via Model Extraction
• Rule Extraction Algorithm for Deep Neural Networks: A Review
• Automatic Rule Extraction from Long Short Term Memory Networks
• Defining Explanation in Probabilistic Systems
• Graying the black box: Understanding DQNs
• Interpreting Black Box Predictions using Fisher Kernels
• Using Visual Analytics to Interpret Predictive Machine Learning Models
• TreeView: Peeking into Deep Neural Networks Via Feature-Space Partitioning
• Towards Transparent AI Systems: Interpreting Visual Question Answering Models
• Visual Explanation by Interpretation: Improving Visual Feedback Capabilities of Deep Neural Networks
• Tree Space Prototypes: Another Look at Making Tree Ensembles Interpretable
• Supersparse Linear Integer Models for Interpretable Classification
• Interpretable Two-level Boolean Rule Learning for Classification
• Local Explanation Methods for Deep Neural Networks Lack Sensitivity to Parameter Values
• InterpNET: Neural Introspection for Interpretable Deep Learning
• Deep saliency: What is learnt by a deep network about saliency?
• "I know it when I see it". Visualization and Intuitive Interpretability

Cited by in corpus (13)

• Interpretable Machine Learning -- A Brief History, State-of-the-Art and Challenges
• Review: Deep Learning in Electron Microscopy
• Achievements and Challenges in Explaining Deep Learning based Computer-Aided Diagnosis Systems
• Explainable AI and Adoption of Financial Algorithmic Advisors: an Experimental Study
• Knowledge-based XAI through CBR: There is more to explanations than models can tell
• Local Concept Embeddings for Analysis of Concept Distributions in Vision DNN Feature Spaces
• Explanatory Pluralism in Explainable AI
• Conjecturing-Based Discovery of Patterns in Data
• Toward Building Science Discovery Machines
• VeriMedi: Pill Identification using Proxy-based Deep Metric Learning and Exact Solution
• Explainability Requires Interactivity
• A User-Centred Framework for Explainable Artificial Intelligence in Human-Robot Interaction
• How to Explain Neural Networks: an Approximation Perspective