Cited by 1,386Explainable Artificial Intelligence (XAI)Software Testing and Debugging TechniquesAdversarial Robustness in Machine Learning

The DOI matches its Crossref record; 46 of the 53 references it cites were found in a registry or an index; none of the 49 sources its claims cite is checked for support yet.

Explainable AI (XAI): Core Ideas, Techniques, and Solutions

Open access

Dwivedi, Rudresh, Dave, Devam, Naik, Het, Singhal, Smiti, Omer, Rana, Patel, Pankesh, Qian, Bin, Wen, Zhenyu, Shah, Tejal, Morgan, Graham, Ranjan, Rajiv·2023·ACM Computing Surveys

DOI 10.1145/3561048

Open this paper in CoChat — free account Open PDF

Abstract

As our dependence on intelligent machines continues to grow, so does the demand for more transparent and interpretable models. In addition, the ability to explain the model generally is now the gold standard for building trust and deployment of artificial intelligence systems in critical domains. Explainable artificial intelligence (XAI) aims to provide a suite of machine learning techniques that enable human users to understand, appropriately trust, and produce more explainable models. Selecting an appropriate approach for building an XAI-enabled application requires a clear understanding of the core ideas within XAI and the associated programming frameworks. We survey state-of-the-art programming techniques for XAI and present the different phases of XAI in a typical machine learning development process. We classify the various XAI approaches and, using this taxonomy, discuss the key differences among the existing XAI techniques. Furthermore, concrete examples are used to describe these techniques that are mapped to programming frameworks and software toolkits. It is the intention that this survey will help stakeholders in selecting the appropriate approaches, programming frameworks, and software toolkits by comparing them through the lens of the presented taxonomy.

AI snapshot

Written from the paper’s full text.

TL;DR

This survey reviews explainable AI concepts, techniques, and software frameworks, organizing them by model transparency, explanation scope, and model specificity. It maps data, model, feature-based, and example-based explanation methods to practical libraries and toolkits, using illustrative examples.

Key findings

The paper proposes a taxonomy of XAI techniques along white-box vs. black-box, model-specific vs. model-agnostic, and global vs. local dimensions, covering data explainability, model explainability, feature-based methods, example-based methods, toolkits, and distributed learning.
It reports an illustrative bias example: a 2015 Google Images search for “CEO” returned women 11% of the time, compared with a real-world representation of 27%.
Neural-symbolic systems are presented as combining neural learning with symbolic reasoning; the paper cites a Neuro-Symbolic Concept Learner trained on 10% of the original dataset while remaining explainable.
It maps techniques such as LIME, SHAP, PDP, ICE, ALE, anchors, counterfactuals, and integrated gradients to frameworks including ELI5, SHAP, PDPbox, Skater, tf-explain, DiCE, Alibi, InterpretML, What-If Tool, and TensorBoard, while noting a trade-off between interpretability and model performance.

Methods

Design: a comparative survey of state-of-the-art XAI programming techniques, not a primary empirical study; it classifies approaches and compares them through a proposed taxonomy.
Data or sample: illustrative datasets and examples include Breast Cancer Wisconsin, Heart Disease, Iris (4 features and 3 target classes), Fashion MNIST (70,000 grayscale 28×28 images across 10 classes), and an animal price prediction example.
Main analysis: qualitative comparison tables mapping XAI techniques to software libraries and toolkits, plus discussion of ML development phases (understanding and explaining) and implementation design considerations.

Limitations

The paper does not state limitations of the survey itself. It does enumerate limitations of reviewed techniques: PDP assumes feature independence and is realistically limited to about two features; ICE can overcrowd plots and shares correlation concerns; LIME’s local linear approximation may fail with nonlinearity; SHAP can be computationally expensive; global surrogates explain the model rather than the data and may not explain individual predictions; counterfactuals perform worse for multiclass than binary classification.
Evident from the text: no benchmark, user study, or empirical validation of the taxonomy or toolkit comparisons is reported; the examples are illustrative rather than systematic evaluations.
Also evident: some tables and sections in the provided text are incomplete or formatting-corrupted, which limits reproducibility of the comparative claims.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1145/3561048

Cites

46 of 46 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 49 cited sources checked · 0 supported · 0 flagged

In the paper
“However, if these models are trained on biased historical data, then such bias can permeate the entire system negatively impacting the decision outcomes.”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > Detecting bias.
What the sources say
- Not checked yetcites [33]
Load-bearingStrength:plain· not yet calibratedCausal:direct· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passes
In the paper
“They can be used only for a single algorithm class.”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > 3.2 Model-specific Techniques vs Model-agnostic Techniques
What the sources say
- Not checked yetcites [44]
Strength:plain· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passes
In the paper
“Model performance metrics (e.g., model accuracy) do not always exhibit true prediction decisions [9].”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > 1 INTRODUCTION
What the sources say
- Not checked yetcites [9] source
Load-bearingStrength:plain· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passes
In the paper
“With the frequent utilization of complex Deep Learning (DL) architectures, it requires urgent attention to understand the inner workings and to get insights into the outcomes.This is a core motivation of Explainable AI (XAI) [15].”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > 1 INTRODUCTION
What the sources say
- Not checked yetcites [15] source
Load-bearingStrength:plain· not yet calibratedCausal:direct· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passes
None of the sources cited for this claim have been checked yet.
In the paper
“However, the state-of-the-art analysis with respect to existing approaches for building XAI-enabled applications has been investigated to a limited extent.”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > 1 INTRODUCTION
What the sources say
- Not checked yetcites Došilović et al. [13]
- Not checked yetcites Bhatt el al. [7]
- Not checked yetcites Arrieta et al. [4]
Load-bearingStrength:plain· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passes
None of the sources cited for this claim have been checked yet.
In the paper
“More recently, a combination of symbolic and subsymbolic approaches based on artificial neural networks, variously known as neuro-symbolic or neural-symbolic (henceforth NS) systems, has shown significant promise in exploiting the complementary strengths of individual approaches [20,28].”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > 5.3 Neural-Symbolic Approaches
What the sources say
- Not checked yetcites [20] source
- Not checked yetcites [28] source
Strength:plain· not yet calibratedCausal:correlational· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passes

Show all 48 claims

Probably not a substantial claim. 118
It may state two separate claims as one. 54

In the paper
“Global interpretation methods involve an overall analysis of a model and its general behavior.The process of defining variables, their dependency, and interactions goes alongside with the process of assigning importance to these components.”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > Global Interpretation.
What the sources say
- Not checked yetcites [44]
Strength:plain· not yet calibratedCausal:correlational· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passes
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“AutoML has made explainability more important, as the entire data pipeline is convered into a black box (as opposed to just the machine learning model) [27].When using AutoML, a user does not have the ability to engineer or select features, or has visibility into a model's decision-making process.”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > Scientific understanding.
What the sources say
- Not checked yetcites [27]
Strength:plain· not yet calibratedCausal:direct· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passes
It may state two separate claims as one.
In the paper
“They are far more intuitive to understand when compared to PDP [44].”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > 6.3 Individual Conditional Expectation (ICE)
What the sources say
- Not checked yetcites [44]
Strength:plain· not yet calibrated
Tests✓ Asserted: passes✗ Checkable: fails✗ Load-bearing: fails✓ Based: passes
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it.
In the paper
“In fact, deep understanding of cognitive science will most certainly benefit the field of NS-AI particularly for the integration of neuro-symbolic approaches [6] to develop a mature and robust integrated model and consequently, for the design of explainable systems that are closer to how humans understand, think, and communicate.”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > --------→ Streptcoccus pyogenes
What the sources say
- Not checked yetcites [6]
Strength:boosted· not yet calibratedCausal:direct· not yet calibrated
Tests✓ Asserted: passes✗ Checkable: fails✗ Load-bearing: fails✓ Based: passes
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it.
In the paper
“Involves an analysis of individual predictions and decisions made by the model, to clarify why the model suggested a particular course of action.”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > Local Interpretation.
What the sources say
- Not checked yetcites [44]
Strength:plain· not yet calibratedCausal:direct· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passes
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“However, to achieve this explainability, the authors argue the critical need for reasoning to understand the why and how behind a particular outcome [14].”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > 5.3 Neural-Symbolic Approaches
What the sources say
- Not checked yetcites [14]
Load-bearingStrength:boosted· not yet calibrated
Tests✓ Asserted: passes✗ Checkable: fails✓ Load-bearing: passes✓ Based: passes
Probably not a substantial claim. It may be too vague to check against evidence.
In the paper
“The ring-allreduce [10] usually achieves better scaleability due to efficient use of network bandwidth.”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > 5.6 Distributed learning
What the sources say
- Not checked yetcites [10]
Strength:hedged· not yet calibratedCausal:direct· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passes
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“In Asynchronous SGD, the delayed gradients create noise in the global model and delay the convergence speed.”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > 5.6 Distributed learning
What the sources say
- Not checked yetcites [11]
Strength:plain· not yet calibratedCausal:direct· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passes
It may state two separate claims as one.
In the paper
“Ontologies enable reasoning not only to deduce new knowledge and facts but also for causal inference to understand the cause and effect behind the outcomes, such as through transitivity [23].”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > 5.3 Neural-Symbolic Approaches
What the sources say
- Not checked yetcites [23]
Strength:plain· not yet calibratedCausal:direct· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passes
It may state two separate claims as one.
In the paper
“Parameter sever (PS) [37] is most prominent for centralized aggregation of data parallel training.”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > 5.6 Distributed learning
What the sources say
- Not checked yetcites [37]
Strength:plain· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passes
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“Mao et al. [41] present a Neuro-Symbolic Concept Learner, which incorporates symbolic reasoning for better interpretation of visual concepts, and can be trained on less data (10% of the original dataset) while being explainable.”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > 5.3 Neural-Symbolic Approaches
What the sources say
- Not checked yetcites [41]
Strength:plain· not yet calibratedCausal:direct· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passes
It may state two separate claims as one.
In the paper
“DiCE not only provides the most influential feature for the prediction but also recommends the feature modifications needed for the result.”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > 7.6 Sotware Libraries for Example-based XAI Techniques
What the sources say
- Not checked yetcites [45]
Strength:plain· not yet calibratedCausal:direct· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passes
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“DiCE allows input of relative difficulty by specifying 'feature weights'.A higher feature weight means that the feature is difficult to modify than that of others.”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > Diverse Counterfactual Explanations (DiCE) for ML
What the sources say
- Not checked yetcites [45]
Strength:plain· not yet calibratedCausal:direct· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passes
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“An important advantage of SHAP value lies in its transparency and interpretability locally, which implies that each observation can have its own SHAP value set.”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > An example.
What the sources say
- Not checked yetcites [40]
Strength:plain· not yet calibratedCausal:direct· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passes
It may state two separate claims as one.
In the paper
“The library consists of a wide range of algorithms, most of which focus on black box models and local interpretation i.e. interpretation for the prediction to be explained at a particular instance.”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > ALIBI.
What the sources say
- Not checked yetcites [34]
Strength:plain· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passes
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“DiCE [45] is a package that gives the “what-if” explanations for the model prediction.”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > 7.6 Sotware Libraries for Example-based XAI Techniques
What the sources say
- Not checked yetcites [45]
Strength:plain· not yet calibratedCausal:direct· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passes
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“It implements counterfactual explanations that include perturbed features which in turn will lead to the required result.”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > 7.6 Sotware Libraries for Example-based XAI Techniques
What the sources say
- Not checked yetcites [45]
Strength:plain· not yet calibratedCausal:direct· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passes
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“This library comprises different types of explainers depending on data we are dealing with.”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > ALIBI.
What the sources say
- Not checked yetcites [34]
Strength:plain· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passes
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“Alibi [34] is an open source Python library which aims at ML-based model inspection and interpretation.”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > ALIBI.
What the sources say
- Not checked yetcites [34]
Strength:plain· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passes
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“Integrated Gradients, also known as Path-Integrated Gradients is an XAI technique that assign an importance value to each feature of the input using the gradients of the model output [57].”
§ Explainable AI (XAI): Core Ideas, Techniques and Solutions > 7.5 Integrated Gradients
What the sources say
- Not checked yetcites [57]
Strength:plain· not yet calibratedCausal:direct· not yet calibrated
Tests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passes
Probably not a substantial claim. The paper’s conclusions may not depend on it.

135 of 135

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

Outline assigns software toolkits to the wrong section
Minor
The outline says the toolkits are covered in Section 9, whereas Section 9 is about design considerations; software toolkits are in Section 8.
Tree ensembles are classified as both white-box and black-box
Minor
Table 3 lists tree ensembles among white-box models, but Section 5.2 explicitly includes them among black-box models.
Table 5 is described as covering the wrong type of explainability
Minor
Section 5.5 says Table 5 concerns data explainability, but the table caption says it concerns model explainability.
Outline misnumbers the data and model explainability sections
Minor
The outline places data explainability in Section 5 and model explainability in Section 6, but their headings are Sections 4 and 5, respectively.
Intrinsic and post-hoc labels are reversed
Minor
Section 3.1 calls black-box models intrinsic and white-box models post-hoc. The conclusion instead describes white-box models as intrinsically explainable relative to black-box models.
Table 8 conflicts with the stated Skater capabilities
Minor
Table 8 marks Skater as unable to perform feature importance and partial dependence plots, but the Skater description expressly says it can perform both.
Low concave-point values receive opposite interpretations
Minor
The PDP discussion associates lower values of the same feature with a lower chance of malignancy, while the ICE discussion associates them with a higher malignant target value.
Outline gives the wrong conclusion section
Minor
The outline places the conclusion in Section 11, but the conclusion is headed Section 10.
Table 6 is assigned two different subjects
Minor
The text identifies Table 6 as a feature-importance example, while its caption identifies it as a table of distributed-learning frameworks.

9 of 9

Greedy function approximation: A gradient boosting machine
Friedman, J. H. (2001) · The Annals of Statistics · doi:10.1214/aos/1013203451 · Cited by 30,731
Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization
Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.74 · Cited by 24,513
Advances in neural information processing systems 7
(1996) · Computers & Mathematics with Applications · doi:10.1016/s0898-1221(96)90463-0 · Cited by 14,759
Advances in Neural Information Processing Systems 29
Teymur, O., Zygalakis, K., & Calderhead, B. (2016) · Cited by 13,881
From local explanations to global understanding with explainable AI for trees
Lundberg, S. M., Erion, G., Chen, H., DeGrave, A., Prutkin, J. M., Nair, B., Katz, R., Himmelfarb, J., Bansal, N., & Lee, S.-I. (2020) · Nature Machine Intelligence · doi:10.1038/s42256-019-0138-9 · Cited by 10,551
Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI
Barredo Arrieta, A., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., Garcia, S., Gil-Lopez, S., Molina, D., Benjamins, R., et al. (2020) · Information Fusion · doi:10.1016/j.inffus.2019.12.012 · Cited by 10,423
Axiomatic attribution for deep networks
Sundararajan, M., Taly, A., & Yan, Q. (2017) · arXiv · doi:10.48550/arxiv.1703.01365 · Cited by 4,087
Towards A Rigorous Science of Interpretable Machine Learning
Doshi-Velez, F., & Kim, B. (2017) · arXiv · doi:10.48550/arxiv.1702.08608 · Cited by 3,121
Learning Important Features Through Propagating Activation Differences
Shrikumar, A., Greenside, P., & Kundaje, A. (2017) · arXiv · doi:10.48550/arxiv.1704.02685 · Cited by 2,761
Adversarial machine learning at scale
Kurakin, A., Goodfellow, I., & Bengio, S. (2016) · arXiv · doi:10.48550/arxiv.1611.01236 · Cited by 2,325

Flagged first, then by citations

Show all 53 references

7 of the 53 references couldn’t be matched to a known paper.

Research works printed but not linked: 13

David Beckett, Tim Berners-Lee, Eric Prud’hommeaux, and Gavin Carothers. 2014. RDF 1.1 Turtle. World Wide Web Consortium 2014 (2014), 18–31.
Jeffrey Dean Greg S. Corrado Rajat Monga Kai Chen Matthieu Devin Quoc V. Le Mark Z. Mao et al. 2012. Large scale distributed deep networks. In Proceedings of the 25th International Conference on Neural Information Processing Systems (NIPS’12) . 1223–1231.
CEUR Workshop Proceedings Proceedings of the First International Workshop on Comprehensibility and Explanation in AI and ML 2017 co-located with 16th International Conference of the Italian Association for Artificial Intelligence (AI*IA 2017) Bari Italy November 16th and 17th 2017 2071 Derek Doran Sarah Schulz Tarek R. Besold Tarek R. Besold Oliver Kutz What does explainable AI really mean? A new conceptualization of perspectives 2017
Xin He Kaiyong Zhao and Xiaowen Chu. 2019. AutoML: A survey of the state-of-the-art. arxiv:1908.00709 [cs.LG]. · arXiv
Pascal Hitzler, Markus Krötzsch, Bijan Parsia, Peter F. Patel-Schneider, and Sebastian Rudolph. 2009. OWL 2 web ontology language primer. W3C Recommendation 27, 1 (2009), 123.
SNOMED International. 2022. Data Analytics with SNOMED CT. Retrieved June 29, 2022 from https://confluence.ihtsdotools.org/display/DOCANLYT/Data+Analytics+with+SNOMED+CT.
Accenture Labs. 2020. Understanding Machines: Explainable AI. Technical Report. Retrieved September 9, 2022 from https://www.accenture.com/_acnmedia/pdf-85/accenture-understanding-machines-explainable-ai.pdf.
Mu Li David G. Andersen Jun Woo Park Alexander J. Smola Amr Ahmed Vanja Josifovski James Long Eugene J. Shekita and Bor-Yiing Su. 2014. Scaling distributed machine learning with the parameter server. In Proceedings of the 11th USENIX Symposium on Operating Systems Design and Implementation (OSDI’14) . 583–598.
Joanne S. Luciano, Bosse Andersson, Colin Batchelor, Olivier Bodenreider, Tim Clark, Christine K. Denney, Christopher Domarew, et al. 2011. The Translational Medicine Ontology and Knowledge Base: Driving personalized medicine by bridging the gap between bench and bedside. Journal of Biomedical Semantics 2 (2011), 1–21.
Azalia Mirhoseini, Anna Goldie, Hieu Pham, Benoit Steiner, Quoc V. Le, and Jeff Dean. 2018. A hierarchical model for device placement. In Proceedings of the International Conference on Learning Representations.
Azalia Mirhoseini, Hieu Pham, Quoc V. Le, Benoit Steiner, Rasmus Larsen, Yuefeng Zhou, Naveen Kumar, Mohammad Norouzi, Samy Bengio, and Jeff Dean. 2017. Device placement optimization with reinforcement learning. In Proceedings of the International Conference on Machine Learning. 2430–2439.
CEUR Workshop Proceedings Proceedings of the Twelfth International Workshop on Neural-Symbolic Learning and Reasoning NeSy 2017 London UK July 17-18 2017 2003 Kamruzzaman Sarker Ning Xie Derek Doran Michael L. Raymer Pascal Hitzler Tarek R. Besold Artur S. d’Avila Garcez Isaac Noble Explaining trained neural networks with semantic web technologies: First steps 2017
Ajay Thampi. 2020. Interpretable AI, Building Explainable Machine Learning Systems. Manning Publications.

Sorted by citations

A comprehensive review of model compression techniques in machine learning
Dantas, P. V., da Silva, W. S., Cordeiro, L. C., & Carvalho, C. B. (2024) · Applied Intelligence · doi:10.1007/s10489-024-05747-w · Cited by 314
Unveiling the black box: A systematic review of Explainable Artificial Intelligence in medical image analysis
Muhammad, D., & Bendechache, M. (2024) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2024.08.005 · Cited by 325
Explainable artificial intelligence: A survey of needs, techniques, applications, and future direction
Mersha, M. A., Lâm, K. N., Wood, J., AlShami, A. K., & Kalita, J. K. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.128111 · Cited by 346
Black-Box Access is Insufficient for Rigorous AI Audits
Casper, S., Ezell, C., Siegmann, C., Kolt, N., Curtis, T. L., Bucknall, B., Haupt, A., Wei, K., Scheurer, J., Hobbhahn, M., et al. (2024) · The 2024 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3630106.3659037 · Cited by 86
A Survey of Explainable Artificial Intelligence in Healthcare: Concepts, Applications, and Challenges
Mienye, I. D., Obaido, G., Jere, N., Mienye, E., Aruleba, K., Emmanuel, I. D., & Ogbuokiri, B. (2024) · Informatics in Medicine Unlocked · doi:10.1016/j.imu.2024.101587 · Cited by 212

See all 1,386

Authors: Dwivedi, R., Dave, D., Naik, H., Singhal, S., Omer, R., Patel, P., Qian, B., Wen, Z., Shah, T., Morgan, G., & Ranjan, R.
Affiliations: Netaji Subhas University of Technology (IN)
Pandit Deendayal Energy University (IN)
Cardiff University, Cardiff, UK
University of South Carolina (US)
Newcastle University (GB)
ORCID: Dwivedi, Rudresh
Dave, Devam
Naik, Het
Singhal, Smiti
Omer, Rana
Patel, Pankesh
Qian, Bin
Wen, Zhenyu
Shah, Tejal
Morgan, Graham
Ranjan, Rajiv
Venue: ACM Computing Surveys
Publisher: Association for Computing Machinery (ACM)
Published: 2023-01-16
Volume: 55
Issue: 9
Pages: 1-33
ISSN: 0360-0300 (ISSN-L)
1557-7341
DOI: 10.1145/3561048
OpenAlex: W4294559022
Semantic Scholar: 3b8abd466697998c6e17df2cd30f48a7594d795b
Cited by: 1,386
References: 53
Citations by year: 2026: 479 · 2025: 554 · 2024: 284 · 2023: 68
FWCI: 74.65
Type: journal article
Language: en
Open access: green
orca.cardiff.ac.uk
Licence: www.acm.org/publications/policies/copyright_policy#Background (vor)
Locations: ACM Computing Surveys (publishedVersion)
ORCA Online Research @Cardiff (Cardiff University) (submittedVersion)
Topics: Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
Physical Sciences › Computer Science › Software › Software Testing and Debugging Techniques
Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning
Keywords: explainable AI, model interpretability, trust in AI

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-10-06, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-06 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Explainable AI (XAI): Core Ideas, Techniques, and Solutions

Abstract

Overview

AI snapshot

TL;DR

Key findings

Methods

Limitations

Integrity signals

Verification

Claims the paper makes

Self-check

Citations

Cites

Cited by

Related

Metadata

Record

About these checks