Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI)
By Amina Adadi, Mohammed Berrada. 2018
5,195 citations recorded by the source corpus.
At the dawn of the fourth industrial revolution, we are witnessing a fast and widespread adoption of artificial intelligence (AI) in our daily life, which contributes to accelerating the shift towards a more algorithmic society. However, even with such unprecedented advancements, a key impediment to the use of AI-based systems is that they often lack transparency. Indeed, the black-box nature of these systems allows powerful predictions, but it cannot be directly explained. This issue has triggered a new debate on explainable AI (XAI). A research field holds substantial promise for improving trust and transparency of AI-based systems. It is recognized as the sine qua non for AI to continue making steady progress without disruption. This survey provides an entry point for interested researchers and practitioners to learn key aspects of the young and rapidly growing body of research related to XAI. Through the lens of the literature, we review the existing approaches regarding the topic, discuss trends surrounding its sphere, and present major research trajectories.
Subfields: Artificial Intelligence, Health Informatics, Safety, Robustness & Interpretability
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Mohit Bansal
- Percy Liang
- Jure Leskovec
- Yoshua Bengio
- Bernhard Schölkopf
- Q. Liao
- Ana Marasovic
- Noah A. Smith
In Health Informatics
- Gesina Schwalbe
- Bettina Finzel
- Timo Speith
- Josivania Silva Farias
- Ian Covert
- Eoin M. Kenny
- Adrien Bennetot
- Eduardo Almeida Soares
In Safety, Robustness & Interpretability
- Longo Luca
- Giulia Vilone
- Dino Pedreschi
- Q. Liao
- Riccardo Guidotti
- Salvatore Ruggieri
- Fosca Giannotti
- Christoph Molnar
Citations in the corpus
- Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AI (2019, neutral): “One of the last surveys very briefly introduced the idea of algorithm property and trade secrets [14] .”
- The Twin-System Approach as One Generic Solution for XAI: An Overview of ANN-CBR Twins for Explaining Deep Learning (2019, background): “…every major AI/ML conference and workshop has targeted the problem, thematically (e.g., NIPS-16, IJCAI-17, IJCAI/ECAI-18, IJCNN-17, ICCBR-18, ICCBR-19, IJCAI19 and the present workshop), along with the emergence of conferences dedicated solely to it (FAT-ML, FAT*19; see [Adadi and Berrada, 2018]).”
- Neural Prototype Trees for Interpretable Fine-grained Image Recognition (2020, background): “We address this socalled ‘accuracy-interpretability trade-off’ [1, 35] by combining the expressiveness of deep learning with the interpretability of decision trees.”
- This Looks Like That, Because ... Explaining Prototypes for Interpretable Image Recognition (2020, background): “Following this demand on understanding automated decision making, explainable Artificial Intelligence (XAI) has been actively researched [1, 3, 18].”
- ViCE: visual counterfactual explanations for machine learning models (2020, neutral): “Previous works [1–4, 6, 10, 15, 17] provide an overview of methods, opportunities and challenges in this area.”
- Learning Explainable Decision Rules via Maximum Satisfiability (2020, background): “Explainable artificial intelligence (XAI) is a family of meth-ods focusing on the algorithmic transparency and inter-pretability of decision procedures [1], [2].”
- A survey of algorithmic recourse: definitions, formulations, solutions, and prospects (2020, background): “For XAI, see [1, 2, 31, 55, 61, 67, 75, 111, 120, 128], for causal and counterfactual inference, see [63, 76, 131, 145, 146, 159, 185], and for optimization, see [32, 137, 166].”
- Persuasive Explanation of Reasoning Inferences on Dietary Data (2019, background): “Explainable Artificial Intelligence (XAI) generally relates to strategies able to provide human-understandable descriptions of learning algorithms usually perceived as black boxes by users [1].”
- Explainable artificial intelligence: an analytical review (2021, methodology): “The existing literature (Adadi & Berrada, 2018) divides XAI taxonomy by: (cid:129) Scope (local (Bach et al., 2015; Selvaraju et al., 2017) and global (Chen et al., 2019; Pedreschi et al., 2019)).”
- Explanation-Based Human Debugging of NLP Models: A Survey (2021, background): “2020), also referred to as transparent (Adadi and Berrada, 2018), inherently interpretable (Rudin, 2019), or directly interpretable (Arya et al.”
- Human-Centered Explainable AI (XAI): From Algorithms to User Experiences (2021, background): “For interested readers, we suggest several recent papers that provided deeper technical surveys [6, 11, 20, 44].”
- Bias in data‐driven artificial intelligence systems—An introductory survey (2020, background): “Although attempts to tackle interpretable ML have existed for some time (Hoffman & Klein, 2017), there has been an exceptional growth of research literature in the last years with emerging keywords such as explainable AI (Adadi & Berrada, 2018) and black box explanation (Guidotti et al., 2019).”
- Unbox the black-box for the medical explainable AI via multi-modal and multi-centre data fusion: A mini-review, two showcases and beyond (2021, background): “Model-agnostic representations are usually post-hoc that are generally used to explain deep neural networks with interpretable surrogates that can be local or global [2].”
- Current Challenges and Future Opportunities for XAI in Machine Learning-Based Clinical Decision Support Systems: A Systematic Review (2021, background): “Although, giving explanations can increase understanding of and trust in a system [19,46,64], simple explanations can hide undesirable attributes of the system and may mislead users into coming to dangerous or unfounded conclusions that could ultimately be unethical [21].”
- Machine Learning Interpretability: A Survey on Methods and Metrics (2019, methodology): “The main goal is to create a suite of interpretable models and methods that produce more explainable models whilst preserving high predictive performance levels [2].”