On the failings of Shapley values for explainability. (English) Zbl 07885914
Summary: Explainable Artificial Intelligence (XAI) is widely considered to be critical for building trust into the deployment of systems that integrate the use of machine learning (ML) models. For more than two decades Shapley values have been used as the theoretical underpinning for some methods of XAI, being commonly referred to as SHAP scores. Some of these methods of XAI now rank among the most widely used, including in high-risk domains. This paper proves that the existing definitions of SHAP scores will necessarily yield misleading information about the relative importance of features for predictions. The paper identifies a number of ways in which misleading information can be conveyed to human decision makers, and proves that there exist classifiers which will yield such misleading information. Furthermore, the paper offers empirical evidence that such theoretical limitations of SHAP scores are routinely observed in ML classifiers.
MSC:
| 68T37 | Reasoning under uncertainty in the context of artificial intelligence |
References:
| [1] | Abchiche-Mimouni, N.; Amgoud, L.; Zehraoui, F., Explainable ensemble classification model based on argumentation, (AAMAS, 2023), 2367-2369 |
| [2] | Adeoye, J.; Zheng, L.-W.; Thomson, P.; Choi, S.-W.; Su, Y.-X., Explainable ensemble learning model improves identification of candidates for oral cancer screening, Oral Oncol., 136, Article 106278 pp., 2023 |
| [3] | Afchar, D.; Guigue, V.; Hennequin, R., Towards rigorous interpretations: a formalisation of feature attribution, (ICML, 2021), 76-86 |
| [4] | Alabi, R. O.; Almangush, A.; Elmusrati, M.; Leivo, I.; Mäkitie, A. A., An interpretable machine learning prognostic system for risk stratification in oropharyngeal cancer, Int. J. Med. Inform., 168, Article 104896 pp., 2022 |
| [5] | Alam, M. S.; Xie, Y., Appley: approximate Shapley value for model explainability in linear time, (Big Data, 2022), 95-100 |
| [6] | Albini, E.; Long, J.; Dervovic, D.; Magazzeni, D., Counterfactual Shapley additive explanations, (FACCT, 2022), 1054-1070 |
| [7] | Ali, S.; Abuhmed, T.; El-Sappagh, S. H.A.; Muhammad, K.; Alonso-Moral, J. M.; Confalonieri, R.; Guidotti, R.; Ser, J. D.; Rodríguez, N. D.; Herrera, F., Explainable artificial intelligence (XAI): what we know and what is left to attain trustworthy artificial intelligence, Inf. Fusion, 99, Article 101805 pp., 2023 |
| [8] | Alsinglawi, B.; Alshari, O.; Alorjani, M.; Mubin, O.; Alnajjar, F.; Novoa, M.; Darwish, O., An explainable machine learning framework for lung cancer hospital length of stay prediction, Sci. Rep., 12, 1-10, 2022 |
| [9] | Amgoud, L., Explaining black-box classification models with arguments, (ICTAI, 2021), 791-795 |
| [10] | Amgoud, L., Non-monotonic explanation functions, (ECSQARU, 2021), 19-31 · Zbl 07542330 |
| [11] | Amgoud, L., Explaining black-box classifiers: properties and functions, Int. J. Approx. Reason., 155, 40-65, 2023 · Zbl 07698204 |
| [12] | Amgoud, L.; Ben-Naim, J., Axiomatic foundations of explainability, (IJCAI, 2022), 636-642 |
| [13] | Amgoud, L.; Muller, P.; Trenquier, H., Argument-based explanation functions, (AAMAS, 2023), 2373-2375 |
| [14] | Amgoud, L.; Muller, P.; Trenquier, H., Leveraging argumentation for generating robust sample-based explanations, (IJCAI, 2023), 3104-3111 |
| [15] | Arenas, M.; Báez, D.; Barceló, P.; Pérez, J.; Subercaseaux, B., Foundations of symbolic languages for model interpretability, (NeurIPS, 2021), 11690-11701 |
| [16] | Arenas, M.; Barceló, P.; Bertossi, L. E.; Monet, M., The tractability of SHAP-score-based explanations for classification over deterministic and decomposable Boolean circuits, (AAAI, 2021), 6670-6678 |
| [17] | Arenas, M.; Barceló, P.; Bertossi, L. E.; Monet, M., On the complexity of SHAP-score-based explanations: tractability via knowledge compilation and non-approximability results, J. Mach. Learn. Res., 24, Article 63 pp., 2023 |
| [18] | Arenas, M.; Barceló, P.; Orth, M. A.R.; Subercaseaux, B., On computing probabilistic explanations for decision trees, (NeurIPS, 2022) |
| [19] | Arrieta, A. B.; Rodríguez, N. D.; Ser, J. D.; Bennetot, A.; Tabik, S.; Barbado, A.; García, S.; Gil-Lopez, S.; Molina, D.; Benjamins, R.; Chatila, R.; Herrera, F., Explainable artificial intelligence (XAI): concepts, taxonomies, opportunities and challenges toward responsible AI, Inf. Fusion, 58, 82-115, 2020 |
| [20] | Audemard, G.; Bellart, S.; Bounia, L.; Koriche, F.; Lagniez, J.; Marquis, P., On the computational intelligibility of Boolean classifiers, (KR, 2021), 74-86 |
| [21] | Audemard, G.; Bellart, S.; Bounia, L.; Koriche, F.; Lagniez, J.; Marquis, P., On preferred abductive explanations for decision trees and random forests, (IJCAI, 2022), 643-650 |
| [22] | Audemard, G.; Bellart, S.; Bounia, L.; Koriche, F.; Lagniez, J.; Marquis, P., Trading complexity for sparsity in random forest explanations, (AAAI, 2022), 5461-5469 |
| [23] | Audemard, G.; Bellart, S.; Lagniez, J.; Marquis, P., Computing abductive explanations for boosted regression trees, (IJCAI, 2023), 3432-3441 |
| [24] | Audemard, G.; Koriche, F.; Marquis, P., On tractable XAI queries based on compiled representations, (KR, 2020), 838-849 |
| [25] | Audemard, G.; Lagniez, J.; Marquis, P.; Szczepanski, N., Computing abductive explanations for boosted trees, (AISTATS, 2023), 4699-4711 |
| [26] | Audemard, G.; Lagniez, J.; Marquis, P.; Szczepanski, N., On contrastive explanations for tree-based classifiers, (ECAI, 2023), 117-124 |
| [27] | Baptista, M. L.; Goebel, K.; Henriques, E. M., Relation between prognostics predictor evaluation metrics and local interpretability SHAP values, Artif. Intell., 306, Article 103667 pp., 2022 · Zbl 07505983 |
| [28] | Bassan, S.; Katz, G., Towards formal XAI: formally approximate minimal explanations of neural networks, (TACAS, 2023), 187-207 · Zbl 07777305 |
| [29] | Béjar, R.; Morgado, A.; Planes, J.; Marques-Silva, J., On logic-based explainability with partially specified inputs, 2023, CoRR |
| [30] | Biradar, G.; Izza, Y.; Lobo, E.; Viswanathan, V.; Zick, Y., Axiomatic aggregations of abductive explanations, 2023, CoRR |
| [31] | Blake, A., Canonical expressions in Boolean algebra, 1937, The University of Chicago, Ph.D. thesis |
| [32] | Blumer, A.; Ehrenfeucht, A.; Haussler, D.; Warmuth, M. K., Occam’s razor, Inf. Process. Lett., 24, 377-380, 1987 · Zbl 0653.68084 |
| [33] | Bodria, F.; Rinzivillo, S.; Fadda, D.; Guidotti, R.; Giannotti, F.; Pedreschi, D., Explaining black box with visual exploration of latent space, (EuroVis - Short Papers, 2022) |
| [34] | Boumazouza, R.; Alili, F. C.; Mazure, B.; Tabia, K., A symbolic approach for counterfactual explanations, (SUM, 2020), 270-277 · Zbl 1517.68325 |
| [35] | Boumazouza, R.; Alili, F. C.; Mazure, B.; Tabia, K., ASTERYX: a model-agnostic sat-based approach for symbolic and score-based explanations, (CIKM, 2021), 120-129 |
| [36] | Boumazouza, R.; Alili, F. C.; Mazure, B.; Tabia, K., Symbolic explanations for multi-label classification, (ICAART, 2023), 342-349 |
| [37] | Brown, F. M., Boolean Reasoning: the Logic of Boolean Equations, 1990, Springer · Zbl 0719.03002 |
| [38] | Camburu, O.; Giunchiglia, E.; Foerster, J. N.; Lukasiewicz, T.; Blunsom, P., The struggles of feature-based explanations: Shapley values vs. minimal sufficient subsets, 2020, CoRR |
| [39] | Campbell, T. W.; Roder, H.; Georgantas, R. W.; Roder, J., Exact Shapley values for local and model-true explanations of decision tree ensembles, Mach. Learn. Appl., 9, Article 100345 pp., 2022 |
| [40] | Carbonnel, C.; Cooper, M. C.; Marques-Silva, J., Tractable explaining of multivariate decision trees, (KR, 2023), 127-135 |
| [41] | Chen, J.; Song, L.; Wainwright, M. J.; Jordan, M. I., L-Shapley and C-Shapley: efficient model interpretation for structured data, (ICLR, 2019) |
| [42] | de Colnet, A.; Marquis, P., On the complexity of enumerating prime implicants from decision-DNNF circuits, (IJCAI, 2022), 2583-2590 |
| [43] | Cooper, M.; Amgoud, L., Abductive explanations of classifiers under constraints: complexity and properties, (ECAI, 2023), 469-476 |
| [44] | Cooper, M. C.; Marques-Silva, J., On the tractability of explaining decisions of classifiers, (CP, 2021), Article 21 pp. |
| [45] | Cooper, M. C.; Marques-Silva, J., Tractability of explaining classifier decisions, Artif. Intell., 316, Article 103841 pp., 2023 · Zbl 07697281 |
| [46] | Covert, I.; Lee, S., Improving KernelSHAP: practical Shapley value estimation using linear regression, (AISTATS, 2021), 3457-3465 |
| [47] | Covert, I.; Lundberg, S. M.; Lee, S., Understanding global feature contributions with additive importance measures, (NeurIPS, 2020) |
| [48] | Crama, Y.; Hammer, P. L., Boolean Functions - Theory, Algorithms, and Applications, Encyclopedia of Mathematics and Its Applications, vol. 142, 2011, Cambridge University Press · Zbl 1237.06001 |
| [49] | Darwiche, A., Three modern roles for logic in AI, (PODS, 2020), 229-243 |
| [50] | Darwiche, A., Logic for explainable AI, (LICS, 2023), 1-11 |
| [51] | Darwiche, A.; Hirth, A., On the reasons behind decisions, (ECAI, 2020), 712-720 · Zbl 1464.68307 |
| [52] | Darwiche, A.; Hirth, A., On the (complete) reasons behind decisions, J. Log. Lang. Inf., 32, 63-88, 2023 · Zbl 07727820 |
| [53] | Darwiche, A.; Ji, C., On the computation of necessary and sufficient explanations, (AAAI, 2022), 5582-5591 |
| [54] | Darwiche, A.; Marquis, P., A knowledge compilation map, J. Artif. Intell. Res., 17, 229-264, 2002 · Zbl 1045.68131 |
| [55] | Darwiche, A.; Marquis, P., On quantifying literals in Boolean logic and its applications to explainable AI, J. Artif. Intell. Res., 72, 285-328, 2021 · Zbl 1522.68526 |
| [56] | Darwiche, A.; Marquis, P., On quantifying literals in Boolean logic and its applications to explainable AI (extended abstract), (IJCAI, 2022), 5718-5721 |
| [57] | Datta, A.; Sen, S.; Zick, Y., Algorithmic transparency via quantitative input influence: theory and experiments with learning systems, (IEEE S&P, 2016), 598-617 |
| [58] | Demšar, J.; Curk, T.; Erjavec, A.; Gorup, Črt; Hočevar, T.; Milutinovič, M.; Možina, M.; Polajnar, M.; Toplak, M.; Starič, A.; Štajdohar, M.; Umek, L.; Žagar, L.; Žbontar, J.; Žitnik, M.; Zupan, B., Orange: data mining toolbox in python, J. Mach. Learn. Res., 14, 2349-2353, 2013 · Zbl 1317.68151 |
| [59] | Dillig, I.; Dillig, T.; McMillan, K. L.; Aiken, A., Minimum satisfying assignments for SMT, (CAV, 2012), 394-409 |
| [60] | Eiter, T.; Gottlob, G., The complexity of logic-based abduction, J. ACM, 42, 3-42, 1995 · Zbl 0886.68121 |
| [61] | Ekin, O.; Foldes, S.; Hammer, P. L.; Hellerstein, L., Equational characterizations of Boolean function classes, Discrete Math., 211, 27-51, 2000 · Zbl 0947.06008 |
| [62] | Friedrich, G.; Gottlob, G.; Nejdl, W., Hypothesis classification, abductive diagnosis and therapy, (ESE, 1990), 69-78 |
| [63] | Frye, C.; de Mijolla, D.; Begley, T.; Cowton, L.; Stanley, M.; Feige, I., Shapley explainability on the data manifold, (ICLR, 2021) |
| [64] | Frye, C.; Rowat, C.; Feige, I., Asymmetric Shapley values: incorporating causal knowledge into model-agnostic explainability, (NeurIPS, 2020) |
| [65] | Fryer, D. V.; Strümke, I.; Nguyen, H. D., Shapley values for feature selection: the good, the bad, and the axioms, IEEE Access, 9, 144352-144360, 2021 |
| [66] | Galazzo, I. B.; Cruciani, F.; Brusini, L.; Salih, A. M.A.; Radeva, P.; Storti, S. F.; Menegaz, G., Explainable artificial intelligence for magnetic resonance imaging aging brainprints: grounds and challenges, IEEE Signal Process. Mag., 39, 99-116, 2022 |
| [67] | Gandolfi, M.; Galazzo, I. B.; Pavan, R. G.; Cruciani, F.; Valè, N.; Picelli, A.; Storti, S. F.; Smania, N.; Menegaz, G., eXplainable AI allows predicting upper limb rehabilitation outcomes in sub-acute stroke patients, IEEE J. Biomed. Health Inform., 27, 263-273, 2023 |
| [68] | Goodfellow, I. J.; Shlens, J.; Szegedy, C., Explaining and harnessing adversarial examples, (ICLR, 2015) |
| [69] | Gorji, N.; Rubin, S., Sufficient reasons for classifier decisions in the presence of domain constraints, (AAAI, 2022), 5660-5667 |
| [70] | Guidotti, R.; Monreale, A.; Ruggieri, S.; Naretto, F.; Turini, F.; Pedreschi, D.; Giannotti, F., Stable and actionable explanations of black-box models through factual and counterfactual rules, Data Min. Knowl. Discov., 1-38, 2022 |
| [71] | Guidotti, R.; Monreale, A.; Ruggieri, S.; Turini, F.; Giannotti, F.; Pedreschi, D., A survey of methods for explaining black box models, ACM Comput. Surv., 51, Article 93 pp., 2019 |
| [72] | Hight, S. L., Minimal input solutions, IEEE Trans. Comput., 20, 923-925, 1971 |
| [73] | Huang, T.; Le, D.; Yuan, L.; Xu, S.; Peng, X., Machine learning for prediction of in-hospital mortality in lung cancer patients admitted to intensive care unit, PLoS ONE, 18, Article e0280606 pp., 2023 |
| [74] | Huang, X.; Cooper, M. C.; Morgado, A.; Planes, J.; Marques-Silva, J., Feature necessity & relevancy in ML classifier explanations, (TACAS, 2023), 167-186 · Zbl 07777304 |
| [75] | Huang, X.; Izza, Y.; Ignatiev, A.; Cooper, M. C.; Asher, N.; Marques-Silva, J., Tractable explanations for d-DNNF classifiers, (AAAI, 2022), 5719-5728 |
| [76] | Huang, X.; Izza, Y.; Ignatiev, A.; Marques-Silva, J., On efficiently explaining graph-based classifiers, (KR, 2021), 356-367 |
| [77] | Huang, X.; Izza, Y.; Marques-Silva, J., Solving explainability queries with quantification: the case of feature relevancy, (AAAI, 2023), 3996-4006 |
| [78] | Huang, X.; Marques-Silva, J., From decision trees to explained decision sets, (ECAI, 2023), 1100-1108 |
| [79] | Huang, X.; Marques-Silva, J., From robustness to explainability and back again, 2023, CoRR |
| [80] | Huang, X.; Marques-Silva, J., The inadequacy of Shapley values for explainability, 2023, CoRR |
| [81] | Huang, X.; Marques-Silva, J., A refutation of Shapley values for explainability, 2023, CoRR |
| [82] | Huang, X.; Marques-Silva, J., Refutation of Shapley values for XAI - additional evidence, 2023, CoRR |
| [83] | Hurault, A.; Marques-Silva, J., Certified logic-based explainable AI - the case of monotonic classifiers, (TAP, 2023), 51-67 · Zbl 1541.68219 |
| [84] | Ignatiev, A., Towards trustable explainable AI, (IJCAI, 2020), 5154-5158 |
| [85] | Ignatiev, A.; Cooper, M. C.; Siala, M.; Hebrard, E.; Marques-Silva, J., Towards formal fairness in azmachine learning, (CP, 2020), 846-867 |
| [86] | Ignatiev, A.; Izza, Y.; Stuckey, P. J.; Marques-Silva, J., Using MaxSAT for efficient explanations of tree ensembles, (AAAI, 2022), 3776-3785 |
| [87] | Ignatiev, A.; Marques-Silva, J., SAT-based rigorous explanations for decision lists, (SAT, 2021), 251-269 · Zbl 07495578 |
| [88] | Ignatiev, A.; Narodytska, N.; Asher, N.; Marques-Silva, J., From contrastive to abductive explanations and back again, (AIxIA, 2020), 335-355 |
| [89] | Ignatiev, A.; Narodytska, N.; Marques-Silva, J., Abduction-based explanations for machine learning models, (AAAI, 2019), 1511-1519 |
| [90] | Ignatiev, A.; Narodytska, N.; Marques-Silva, J., On relating explanations and adversarial examples, (NeurIPS, 2019), 15857-15867 |
| [91] | Ignatiev, A.; Narodytska, N.; Marques-Silva, J., On validating, repairing and refining heuristic ML explanations, 2019, CoRR |
| [92] | Inoguchi, T.; Nohara, Y.; Nojiri, C.; Nakashima, N., Association of serum bilirubin levels with risk of cancer development and total death, Sci. Rep., 11, 1-12, 2021 |
| [93] | Izza, Y.; Huang, X.; Ignatiev, A.; Narodytska, N.; Cooper, M. C.; Marques-Silva, J., On computing probabilistic abductive explanations, Int. J. Approx. Reason., 159, Article 108939 pp., 2023 · Zbl 07713736 |
| [94] | Izza, Y.; Ignatiev, A.; Marques-Silva, J., On tackling explanation redundancy in decision trees, J. Artif. Intell. Res., 75, 261-321, 2022 · Zbl 07603113 |
| [95] | Izza, Y.; Ignatiev, A.; Marques-Silva, J., On tackling explanation redundancy in decision trees (extended abstract), (IJCAI, 2023), 6900-6904 |
| [96] | Izza, Y.; Ignatiev, A.; Stuckey, P. J.; Marques-Silva, J., Delivering inflated explanations, 2023, CoRR |
| [97] | Izza, Y.; Marques-Silva, J., On explaining random forests with SAT, (IJCAI, 2021), 2584-2591 |
| [98] | Jansen, T.; Geleijnse, G.; Van Maaren, M.; Hendriks, M. P.; Ten Teije, A.; Moncada-Torres, A., Machine learning explainability in breast cancer survival, (Digital Personalized Health and Medicine, 2020, IOS Press), 307-311 |
| [99] | Jethani, N.; Sudarshan, M.; Covert, I. C.; Lee, S.; Ranganath, R., FastSHAP: real-time Shapley value estimation, (ICLR, 2022) |
| [100] | Ji, C.; Darwiche, A., A new class of explanations for classifiers with non-binary features, (JELIA, 2023), 106-122 · Zbl 07855115 |
| [101] | Kam, T. Y.-k.; Brayton, R. K., Multi-Valued Decision Diagrams, 1990, University of California Berkeley, Technical Report UCB/ERL M90/125 |
| [102] | Kumar, I.; Scheidegger, C.; Venkatasubramanian, S.; Friedler, S. A., Shapley residuals: quantifying the limits of the Shapley value for explanations, (NeurIPS, 2021), 26598-26608 |
| [103] | Kumar, I. E.; Venkatasubramanian, S.; Scheidegger, C.; Friedler, S. A., Problems with Shapley-value-based explanations as feature importance measures, (ICML, 2020), 5491-5500 |
| [104] | Labreuche, C., Explanation of pseudo-Boolean functions using cooperative game theory and prime implicants, (SUM, 2022), 295-308 · Zbl 1524.68252 |
| [105] | Ladbury, C.; Li, R.; Shiao, J.; Liu, J.; Cristea, M.; Han, E.; Dellinger, T.; Lee, S.; Wang, E.; Fisher, C., Characterizing impact of positive lymph node number in endometrial cancer using machine-learning: a better prognostic indicator than FIGO staging?, Gynecol. Oncol., 164, 39-45, 2022 |
| [106] | Lelis, V. M.; Guzmán, E.; Belmonte, M., Non-invasive meningitis diagnosis using decision trees, IEEE Access, 8, 18394-18407, 2020 |
| [107] | Lipovetsky, S.; Conklin, M., Analysis of regression in game theory approach, Appl. Stoch. Models Bus. Ind., 17, 319-330, 2001 · Zbl 1008.62041 |
| [108] | Liu, X.; Lorini, E., A logic for binary classifiers and their explanation, (CLAR, 2021), 302-321 · Zbl 1541.03109 |
| [109] | Liu, X.; Lorini, E., A logic of “black box” classifier systems, (WoLLIC, 2022), 158-174 · Zbl 1541.68360 |
| [110] | Liu, X.; Lorini, E., A unified logical framework for explanations in classifier systems, J. Log. Comput., 33, 485-515, 2023 · Zbl 07669828 |
| [111] | Liu, Y.; Liu, Z.; Luo, X.; Zhao, H., Diagnosis of Parkinson’s disease based on SHAP value feature selection, Biocybern. Biomed. Eng., 42, 856-869, 2022 |
| [112] | Loh, H. W.; Ooi, C. P.; Seoni, S.; Barua, P. D.; Molinari, F.; Acharya, U. R., Application of explainable artificial intelligence for healthcare: a systematic review of the last decade (2011-2022), Comput. Methods Programs Biomed., 226, Article 107161 pp., 2022 |
| [113] | Lundberg, S. M.; Erion, G. G.; Chen, H.; DeGrave, A. J.; Prutkin, J. M.; Nair, B.; Katz, R.; Himmelfarb, J.; Bansal, N.; Lee, S., From local explanations to global understanding with explainable AI for trees, Nat. Mach. Intell., 2, 56-67, 2020 |
| [114] | Lundberg, S. M.; Lee, S., A unified approach to interpreting model predictions, (NeurIPS, 2017), 4765-4774 |
| [115] | Ma, M.; Liu, R.; Wen, C.; Xu, W.; Xu, Z.; Wang, S.; Wu, J.; Pan, D.; Zheng, B.; Qin, G., Predicting the molecular subtype of breast cancer and identifying interpretable imaging features using machine learning algorithms, Eur. Radiol., 1-11, 2022 |
| [116] | Malfa, E. L.; Michelmore, R.; Zbrzezny, A. M.; Paoletti, N.; Kwiatkowska, M., On guaranteed optimal robust explanations for NLP models, (IJCAI, 2021), 2658-2665 |
| [117] | Marques-Silva, J., Logic-based explainability in machine learning, 2022, CoRR |
| [118] | Marques-Silva, J., Logic-based explainability in machine learning, (Reasoning Web, 2022), 24-104 |
| [119] | Marques-Silva, J.; Gerspacher, T.; Cooper, M. C.; Ignatiev, A.; Narodytska, N., Explaining naive Bayes and other linear classifiers with polynomial time and delay, (NeurIPS, 2020) |
| [120] | Marques-Silva, J.; Gerspacher, T.; Cooper, M. C.; Ignatiev, A.; Narodytska, N., Explanations for monotonic classifiers, (ICML, 2021), 7469-7479 |
| [121] | Marques-Silva, J.; Huang, X., Explainability is NOT a game, 2023, CoRR |
| [122] | Marques-Silva, J.; Ignatiev, A., Delivering trustworthy AI through formal XAI, (AAAI, 2022), 12342-12350 |
| [123] | McCluskey, E. J., Minimization of Boolean functions, Bell Syst. Tech. J., 35, 1417-1444, 1956 |
| [124] | Merrick, L.; Taly, A., The explanation game: explaining machine learning models using Shapley values, (CDMAKE, 2020), 17-38 |
| [125] | Miller, G. A., The magical number seven, plus or minus two: some limits on our capacity for processing information, Psychol. Rev., 63, 81-97, 1956 |
| [126] | Miller, T., Explanation in artificial intelligence: insights from the social sciences, Artif. Intell., 267, 1-38, 2019 · Zbl 1478.68274 |
| [127] | Minh, D.; Wang, H. X.; Li, Y. F.; Nguyen, T. N., Explainable artificial intelligence: a comprehensive review, Artif. Intell. Rev., 55, 3503-3568, 2022 |
| [128] | C. Molnar, Interpretable machine learning, 2020. |
| [129] | Moncada-Torres, A.; van Maaren, M. C.; Hendriks, M. P.; Siesling, S.; Geleijnse, G., Explainable machine learning can outperform Cox regression predictions and provide insights in breast cancer survival, Sci. Rep., 11, 6968, 2021 |
| [130] | Mothilal, R. K.; Mahajan, D.; Tan, C.; Sharma, A., Towards unifying feature attribution and counterfactual explanations: different means to the same end, (AIES, 2021), 652-663 |
| [131] | Nagahisarchoghaei, M.; Nur, N.; Cummins, L.; Nur, N.; Karimi, M. M.; Nandanwar, S.; Bhattacharyya, S.; Rahimi, S., An empirical survey on explainable ai technologies: recent trends, use-cases, and categories from technical and application perspectives, Electronics, 12, 1092, 2023 |
| [132] | Niveau, A.; Fargier, H.; Pralet, C., Representing CSPS with set-labeled diagrams: a compilation map, (Graph Structures for Knowledge Representation and Reasoning: Second International Workshop. Graph Structures for Knowledge Representation and Reasoning: Second International Workshop, GKR 2011, Barcelona, Spain, July 16, 2011, 2011, Springer), 137-171, Revised Selected Papers |
| [133] | Olson, R. S.; La Cava, W.; Orzechowski, P.; Urbanowicz, R. J.; Moore, J. H., PMLB: a large benchmark suite for machine learning evaluation and comparison, BioData Min., 10, 1-13, 2017 |
| [134] | Quine, W. V., The problem of simplifying truth functions, Am. Math. Mon., 59, 521-531, 1952 · Zbl 0048.24503 |
| [135] | Quine, W. V., A way to simplify truth functions, Am. Math. Mon., 62, 627-631, 1955 · Zbl 0068.24209 |
| [136] | Quine, W. V., On cores and prime implicants of truth functions, Am. Math. Mon., 66, 755-760, 1959 · Zbl 0201.32203 |
| [137] | Reiter, R., A theory of diagnosis from first principles, Artif. Intell., 32, 57-95, 1987 · Zbl 0643.68122 |
| [138] | Ribeiro, M. T.; Singh, S.; Guestrin, C., “Why should I trust you?”: explaining the predictions of any classifier, (KDD, 2016), 1135-1144 |
| [139] | Ribeiro, M. T.; Singh, S.; Guestrin, C., Anchors: high-precision model-agnostic explanations, (AAAI, 2018), 1527-1535 |
| [140] | Rodríguez, N. D.; Lamas, A.; Sanchez, J.; Franchi, G.; Donadello, I.; Tabik, S.; Filliat, D.; Cruz, P.; Montes, R.; Herrera, F., Explainable neural-symbolic learning (X-NeSyL) methodology to fuse deep learning representations with expert knowledge graphs: the MonuMAI cultural heritage use case, Inf. Fusion, 79, 58-83, 2022 |
| [141] | Roth, A. E., The Shapley Value: Essays in Honor of Lloyd S. Shapley, 1988, Cambridge University Press |
| [142] | Rozemberczki, B.; Watson, L.; Bayer, P.; Yang, H.; Kiss, O.; Nilsson, S.; Sarkar, R., The Shapley value in machine learning, (IJCAI, 2022), 5572-5579 |
| [143] | Rudin, C., Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead, Nat. Mach. Intell., 1, 206-215, 2019 |
| [144] | Rudin, C.; Chen, C.; Chen, Z.; Huang, H.; Semenova, L.; Zhong, C., Interpretable machine learning: fundamental principles and 10 grand challenges, Stat. Surv., 16, 1-85, 2022 · Zbl 07471610 |
| [145] | (Samek, W.; Montavon, G.; Vedaldi, A.; Hansen, L. K.; Müller, K., Explainable AI: Interpreting, Explaining and Visualizing Deep Learning, 2019, Springer) |
| [146] | Sarvmaili, M.; Guidotti, R.; Monreale, A.; Soares, A.; Sadeghi, Z.; Giannotti, F.; Pedreschi, D.; Matwin, S., A modularized framework for explaining black box classifiers for text data, (CCAI, 2022) |
| [147] | Selman, B.; Levesque, H. J., Abductive and default reasoning: a computational core, (AAAI, 1990), 343-348 |
| [148] | Ser, J. D.; Arrieta, A. B.; Rodríguez, N. D.; Herrera, F.; Saranti, A.; Holzinger, A., On generating trustworthy counterfactual explanations, Inf. Sci., 655, Article 119898 pp., 2024 |
| [149] | Shap, Shap impact, 2023 |
| [150] | Shapley, L. S., A value for n-person games, Contrib. Theor. Games, 2, 307-317, 1953 · Zbl 0050.14404 |
| [151] | Shih, A.; Choi, A.; Darwiche, A., A symbolic approach to explaining Bayesian network classifiers, (IJCAI, 2018), 5103-5111 |
| [152] | Shih, A.; Choi, A.; Darwiche, A., Compiling Bayesian network classifiers into decision graphs, (AAAI, 2019), 7966-7974 |
| [153] | Slack, D.; Hilgard, A.; Singh, S.; Lakkaraju, H., Reliable post hoc explanations: modeling uncertainty in explainability, (NeurIPS, 2021), 9391-9404 |
| [154] | Sorayaie Azar, A.; Babaei Rikan, S.; Naemi, A.; Bagherzadeh Mohasefi, J.; Pirnejad, H.; Bagherzadeh Mohasefi, M.; Wiil, U. K., Application of machine learning techniques for predicting survival in ovarian cancer, BMC Med. Inform. Decis. Mak., 22, 345, 2022 |
| [155] | Strumbelj, E.; Kononenko, I., An efficient explanation of individual classifications using game theory, J. Mach. Learn. Res., 11, 1-18, 2010 · Zbl 1242.68250 |
| [156] | Strumbelj, E.; Kononenko, I., Explaining prediction models and individual predictions with feature contributions, Knowl. Inf. Syst., 41, 647-665, 2014 |
| [157] | Sundararajan, M.; Najmi, A., The many Shapley values for model explanation, (ICML, 2020), 9269-9278 |
| [158] | Van den Broeck, G.; Lykov, A.; Schleich, M.; Suciu, D., On the tractability of SHAP explanations, (AAAI, 2021), 6505-6513 |
| [159] | Van den Broeck, G.; Lykov, A.; Schleich, M.; Suciu, D., On the tractability of SHAP explanations, J. Artif. Intell. Res., 74, 851-886, 2022 · Zbl 07566000 |
| [160] | Vilone, G.; Longo, L., Notions of explainability and evaluation approaches for explainable artificial intelligence, Inf. Fusion, 76, 89-106, 2021 |
| [161] | Voukelatou, V.; Miliou, I.; Giannotti, F.; Pappalardo, L., Understanding peace through the world news, EPJ Data Sci., 11, Article 2 pp., 2022 |
| [162] | Wäldchen, S.; MacDonald, J.; Hauch, S.; Kutyniok, G., The computational complexity of understanding binary classifier decisions, J. Artif. Intell. Res., 70, 351-387, 2021 · Zbl 1512.68116 |
| [163] | Wang, Y.; Lang, J.; Zuo, J. Z.; Dong, Y.; Hu, Z.; Xu, X.; Zhang, Y.; Wang, Q.; Yang, L.; Wong, S. T., The radiomic-clinical model using the SHAP method for assessing the treatment response of whole-brain radiotherapy: a multicentric study, Eur. Radiol., 1-11, 2022 |
| [164] | Watson, D. S., Rational Shapley values, (FAccT, 2022), 1083-1094 |
| [165] | Watson, D. S.; Gultchin, L.; Taly, A.; Floridi, L., Local explanations via necessity and sufficiency: unifying theory and practice, (UAI, vol. 161, 2021), 1382-1392 |
| [166] | Winter, E., The Shapley value, (Handbook of Game Theory with Economic Applications, vol. 3, 2002), 2025-2054 |
| [167] | Withnell, E.; Zhang, X.; Sun, K.; Guo, Y., Xomivae: an interpretable deep learning model for cancer classification using high-dimensional omics data, Brief. Bioinform., 22, Article bbab315 pp., 2021 |
| [168] | Wu, M.; Wu, H.; Barrett, C. W., VeriX: towards verified explainability of deep neural networks, 2022, CoRR |
| [169] | Yan, T.; Procaccia, A. D., If you like Shapley then you’ll love the core, (AAAI, 2021), 5751-5759 |
| [170] | Young, K.; Booth, G.; Simpson, B.; Dutton, R.; Shrapnel, S., Deep neural network or dermatologist?, 2019, CoRR |
| [171] | Yu, D.; Liu, Z.; Su, C.; Han, Y.; Duan, X.; Zhang, R.; Liu, X.; Yang, Y.; Xu, S., Copy number variation in plasma as a tool for lung cancer prediction using extreme gradient boosting (XGBoost) classifier, Thorac. Cancer, 11, 95-102, 2020 |
| [172] | Yu, J.; Farr, G.; Ignatiev, A.; Stuckey, P. J., Anytime approximate formal feature attribution, 2023, CoRR |
| [173] | Yu, J.; Ignatiev, A.; Stuckey, P. J., From formal boosted tree explanations to interpretable rule sets, (CP, 2023), Article 38 pp. |
| [174] | Yu, J.; Ignatiev, A.; Stuckey, P. J., On formal feature attribution and its approximation, 2023, CoRR |
| [175] | Yu, J.; Ignatiev, A.; Stuckey, P. J.; Narodytska, N.; Marques-Silva, J., Eliminating the impossible, whatever remains must be true: on extracting and applying background knowledge in the context of formal explanations, (AAAI, 2023), 4123-4131 |
| [176] | Zarinshenas, R.; Ladbury, C.; McGee, H.; Raz, D.; Erhunmwunsee, L.; Pathak, R.; Glaser, S.; Salgia, R.; Williams, T.; Amini, A., Machine learning to refine prognostic and predictive nodal burden thresholds for post-operative radiotherapy in completely resected stage III-N2 non-small cell lung cancer, Radiother. Oncol., 173, 10-18, 2022 |
| [177] | Zhang, G.; Shi, Y.; Yin, P.; Liu, F.; Fang, Y.; Li, X.; Zhang, Q.; Zhang, Z., A machine learning model based on ultrasound image features to assess the risk of sentinel lymph node metastasis in breast cancer patients: applications of scikit-learn and SHAP, Front. Oncol., 12, 2022 |
| [178] | Zhang, Y.; Weng, Y.; Lund, J., Applications of explainable artificial intelligence in diagnosis and surgery, Diagnostics, 12, 237, 2022 |
| [179] | Zhou, Z.-H., Machine Learning, 2021, Springer Nature · Zbl 1479.68001 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
