Counterfactuals and Causability in Explainable Artificial Intelligence: Theory, Algorithms, and Applications
arXiv:2103.04244
Abstract
There has been a growing interest in model-agnostic methods that can make deep learning models more transparent and explainable to a user. Some researchers recently argued that for a machine to achieve a certain degree of human-level explainability, this machine needs to provide human causally understandable explanations, also known as causability. A specific class of algorithms that have the potential to provide causability are counterfactuals. This paper presents an in-depth systematic review of the diverse existing body of literature on counterfactuals and causability for explainable artificial intelligence. We performed an LDA topic modelling analysis under a PRISMA framework to find the most relevant literature articles. This analysis resulted in a novel taxonomy that considers the grounding theories of the surveyed algorithms, together with their underlying properties and applications in real-world data. This research suggests that current model-agnostic counterfactual algorithms for explainable AI are not grounded on a causal theoretical formalism and, consequently, cannot promote causability to a human decision-maker. Our findings suggest that the explanations derived from major algorithms in the literature provide spurious correlations rather than cause/effects relationships, leading to sub-optimal, erroneous or even biased explanations. This paper also advances the literature with new directions and challenges on promoting causability in model-agnostic approaches for explainable artificial intelligence.
References in corpus (25)
- Towards A Rigorous Science of Interpretable Machine Learning
- Semantics derived automatically from language corpora contain human-like biases
- Interpretable machine learning: definitions, methods, and applications
- European Union regulations on algorithmic decision-making and a "right to explanation"
- Confounding variables can degrade generalization performance of radiological deep learning models
- Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings
- Explaining Machine Learning Classifiers through Diverse Counterfactual Explanations
- Word Embeddings Quantify 100 Years of Gender and Ethnic Stereotypes
- What do we need to build explainable AI systems for the medical domain?
- The What-If Tool: Interactive Probing of Machine Learning Models
- Opportunities and Challenges in Explainable Artificial Intelligence (XAI): A Survey
- Measuring the Quality of Explanations: The System Causability Scale (SCS). Comparing Human and Machine Explanations
- Metrics for Explainable AI: Challenges and Prospects
- The Hidden Assumptions Behind Counterfactual Explanations and Principal Reasons
- Local Rule-Based Explanations of Black Box Decision Systems
- Learning Model-Agnostic Counterfactual Explanations for Tabular Data
- Interpretability via Model Extraction
- Interpretable Credit Application Predictions With Counterfactual Explanations
- PRINCE: Provider-side Interpretability with Counterfactual Explanations in Recommender Systems
- Algorithmic recourse under imperfect causal knowledge: a probabilistic approach
- What is "Intelligent" in Intelligent User Interfaces? A Meta-Analysis of 25 Years of IUI
- Generating Counterfactual and Contrastive Explanations using SHAP
- Outcome-Explorer: A Causality Guided Interactive Visual Interface for Interpretable Algorithmic Decision Making
- Causality Learning: A New Perspective for Interpretable Machine Learning
- Evaluating Explainable Methods for Predictive Process Analytics: A Functionally-Grounded Approach