Accurate intelligible models with pairwise interactions
The DOI matches its Crossref record; 21 of the 22 references it cites were found in a registry or an index.
Abstract
Standard generalized additive models (GAMs) usually model the dependent variable as a sum of univariate models. Although previous studies have shown that standard GAMs can be interpreted by users, their accuracy is significantly less than more complex models that permit interactions.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“It is designed to have high accuracy compared to the state-of-the-art machine learning models while keeping the intelligibility, and explainability [13].”
From An Empirical Comparison of Interpretable Models to Post-Hoc Explanations (2023) · cites this paper as [13]
Citations
Cites
Greedy function approximation: A gradient boosting machine
Friedman, J. H. (2001) · The Annals of Statistics · doi:10.1214/aos/1013203451 · Cited by 30,731
Generalized Additive Models: An Introduction with R
Wood, S. N. (2017) · doi:10.1201/9781315370279 · Cited by 13,852
Introduction to information retrieval
(2009) · Choice Reviews Online · doi:10.5860/choice.46-2715 · Cited by 12,278
Brown, R. A., Hastie, T. J., & Tibshirani, R. J. (1991) · Biometrics · doi:10.2307/2532174 · Cited by 8,176
An introduction to variable and feature selection
GuyonIsabelle, & ElisseeffAndré (2003) · Journal of Machine Learning Research · doi:10.5555/944919.944968 · Cited by 7,715
Koyak, R. A., Hastie, T. J., & Tibshirani, R. J. (1991) · Journal of the American Statistical Association · doi:10.2307/2290538 · Cited by 7,596
Wood, S. N. (2006) · doi:10.1201/9781420010404 · Cited by 4,953
Hastie, T., & Tibshirani, R. (2014) · Wiley StatsRef: Statistics Reference Online · doi:10.1002/9781118445112.stat03141 · Cited by 3,156
An Empirical Comparison of Voting Classification Algorithms: Bagging, Boosting, and Variants
Bauer, E., & Kohavi, R. (1999) · Machine Learning · doi:10.1023/a:1007515423169 · Cited by 2,638
Wood, S. N. (2003) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/1467-9868.00374 · Cited by 2,445
Research works printed but not linked: 2
- 10.5555/1394399 · doi:10.5555/1394399
- 10.5555/1622620.1622633 · doi:10.5555/1622620.1622633
Cited by
Items 301–400 of 565
Byun, J., Lee, J., Lee, H. W., & Son, B. (2024) · IEEE Access · doi:10.1109/access.2024.3523967 · Cited by 6
Jang, J., Kim, M., Bui, T.-C., & Li, W.-S. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-33377-4_13 · Cited by 5
Shrivastava, H., & Chajewska, U. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-47679-2_2 · Cited by 8
Will you Buy it Now?: Predicting Passengers that Purchase Premium Promotions Using the PAX Model
Emadi, N. A., Thirumuruganathan, S., Robillos, D. R., & Jansen, B. J. J. (2021) · Journal of Smart Tourism · doi:10.52255/smarttourism.2021.1.1.7 · Cited by 8
“…(Ribeiro et al., 2016 In Figure 10, we observe an example of an individual prediction of the not-upgraded class by Explainable Boosting Machine (Caruana et al., 2015; Lengerich et al., 2020; Lou et al., 2013, 2012; Tan et al., 2018), with nationality and price data mainly impacting the decision.”
Jeyasothy, A., Ramasamy, S., & Sundaram, S. (2024) · Applied Soft Computing · doi:10.1016/j.asoc.2024.111747 · Cited by 5
Interpretable Learning-to-Rank with Generalized Additive ModelsPreprint
Zhuang, H., Wang, X., Bendersky, M., Grushetsky, A., Wu, Y., Mitrichev, P., Sterling, E., Bell, N., Ravina, W., & Qian, H. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2005.02553 · Cited by 5
“…ranking GAMs: (1) Incorporating limited pairwise feature interactions to improve model accuracy with reasonable interpretability (similar to [31]); (2) Enabling users to dictate constraints such as monotonicity on sub-models; (3) Combining ranking GAMs with fully-fledged neural networks or…”
BackgroundMethodSurrogate Locally-Interpretable Models with Supervised Machine Learning AlgorithmsPreprint
Hu, L., Chen, J., Nair, V. N., & Sudjianto, A. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2007.14528 · Cited by 23
Barbado, A., & Corcho, Ó. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2107.06031 · Cited by 9
“EBM is based on the GA 2 M algorithm [25], but with a difference in terms of computation performance.”
Hecksteden, A., Kempe, M., & Berger, J. (2025) · Science and Medicine in Football · doi:10.1080/24733938.2025.2517056 · Cited by 7
“Thus, as in the original work, we trained a fasterrisk score and an EBM on the training data and apply it to the respective testing data four times in a round-robin procedure.”
Gao, C., Wang, H., Ge, Q., & Dai, J. (2024) · Ecological Indicators · doi:10.1016/j.ecolind.2024.112402 · Cited by 6
Tian, J., & Yang, W. (2024) · Entropy · doi:10.3390/e26110902 · Cited by 6
“Interpretability is domain-specific and frequently constrained by structural knowledge, such as monotonicity [47], causality, additivity [48], or domain-specific physical constraints.”
Galuppini, G., Liang, Q., Tamirisa, P. A., Lemmerman, J. A., Sullivan, M. G., Mazack, M. J. M., Gomadam, P. M., Bazant, M. Z., & Braatz, R. D. (2024) · Journal of Power Sources · doi:10.1016/j.jpowsour.2024.234668 · Cited by 4
“A generalized additive model (GAM) [38,44,52,53] is an inter-pretable model that explains a response variable 𝑦 ∼ ( 𝜇,𝜎 2 ) using a sum of univariate and bivariate (interactions) shape functions of the predictors, where ̂𝑦 is the GAM prediction (corresponding to the expected value of 𝑦 ), 𝑐…”
Explainable Machine Learning Models for Predicting Recurrence in Differentiated Thyroid Cancer
Arslan, A. K., & Çolak, C. (2024) · Medical Records · doi:10.37990/medr.1525801 · Cited by 4
Interpretable machine learning for precision cognitive aging
Mahamadou, A. J. D., Rodrigues, E. A., Vakorin, V. A., Antoine, V., & Moreno, S. (2025) · Frontiers in Computational Neuroscience · doi:10.3389/fncom.2025.1560064 · Cited by 8
Xin, X., Hooker, G., & Huang, F. (2025) · Insurance: Mathematics and Economics · doi:10.1016/j.insmatheco.2025.103135 · Cited by 6
Self-citationRevealing the structure of prediction models through feature interaction detection
Zhang, X., Zhang, H., Zhu, J., & Li, Z. (2021) · Knowledge-Based Systems · doi:10.1016/j.knosys.2021.107737 · Cited by 5
Arslan, A. K., Yağın, F. H., Algarni, A., Al-Hashem, F. H., & Ardigò, L. P. (2024) · Diagnostics · doi:10.3390/diagnostics14131353 · Cited by 7
“The binary interaction terms in the formula are automatically added to the model using the FAST (Feature-based Algorithm for Subgroup Discovery) technique [18].”
MethodGuo, M., Zhang, Q., & Zeng, D. D. (2025) · ACM Transactions on Knowledge Discovery from Data · doi:10.1145/3715150 · Cited by 4
“In this regard, [36] propose GA 2 M , which first learns a base GAM without any interactions and then selects a number of the most important pairwise interactions that minimize residuals.”
Kim, M., Choi, H.-S., & Kim, J. (2025) · Proceedings · doi:10.1109/icdm65498.2025.00140 · Cited by 20
“GA 2 M [20] extends GAM by incorporating second-order (pairwise) feature interactions.”
Wang, P., Padthe, K., Vinzamuri, B., & Reddy, C. K. A. (2016) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/2983323.2983779 · Cited by 4
pureGAM: Learning an Inherently Pure Additive Model
Sun, X., Wang, Z., Ding, R., Han, S., & Zhang, D. (2022) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3534678.3539256 · Cited by 7
“1 INTRODUCTION Including pairwise or higher-order interactions among the predictors (or features) of a Generalized Additive Model (GAM) is gaining increasing attention in the literature [5, 18, 21, 33].”
BackgroundMethodda Silva, V., Papa, J. P., & da Costa, K. A. P. (2023) · International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP) · doi:10.5220/0011664100003417 · Cited by 5
“Secondly, the algorithm selects a fixed number of pairwise interactions using the ranking procedure proposed by [16] and fit their respective subnetworks on the residuals, pruning trivial ones as in the first stage.”
BackgroundMethodInfluentialHuman-AI Collaboration for the Detection of Deceptive Speech
Tutul, A. A., Chaspari, T., Levitan, S. I., & Hirschberg, J. (2023) · International Conference on Affective Computing and Intelligent Interaction Workshops and Demos (ACIIW) · doi:10.1109/aciiw59127.2023.10388114 · Cited by 4
“EBM was trained on 75% of the CXD data using acoustic and linguistic features (Section III-A).”
InfluentialDiscovering Unexpected Local Nonlinear Interactions in Scientific Black-box Models
Doron, M., Segev, I., & Shahaf, D. (2019) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3292500.3330886 · Cited by 4
“The same git repository holds the code running the benchmarks Anova, GA2M and PDP, as well as the synthetic and computational models in Sections 3 and 4 and compressed files containing the data simulated from the computational models.”
BackgroundMethodResultInfluentialEfficiently Training Intelligible Models for Global Explanations
Lou, Y., Wang, Y., Liang, S., & Yang, D. (2020) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3340531.3412702 · Cited by 4
“As suggested by Lou et al. [15], building such single-feature regression tree is equivalent to cutting on a line, which can be made more efficient.”
BackgroundMethodInfluentialSelf-citationImangaliyev, S., Prodan, A., Nieuwdorp, M., Groen, A. K., van Riel, N. A. W., & Levin, E. (2018) · Methods · doi:10.1016/j.ymeth.2018.06.011 · Cited by 4
“have been successfully applied in the biomedical domain [18], and they can be naturally extended to include pairwise interactions among predictors [17].”
BackgroundOn Parameter Interpretability of Phenomenological-Based Semiphysical ModelsPreprint
Lema-Pérez, L., Muñoz‐Tamayo, R., Garcia‐Tirado, J., & Álvarez, H. (2018) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/446583 · Cited by 4
“[1,12] use intelligible models to refer to those models that can be easily interpreted by users.”
BackgroundMethodAn AI-Augmented Lesion Detection Framework For Liver Metastases With Model InterpretabilityPreprint
Hunt, X. J., Abbey, R., Tharrington, R., Huiskens, J., & Wesdorp, N. J. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1907.07713 · Cited by 7
“The majority of methods for model interpretability come in three forms: inherently-interpretable models [3, 15, 27], methods for interpreting existing models [19, 25]), and post-hoc investigations [16, 23, 24, 26].”
MethodDNN2LR: Interpretation-inspired Feature Crossing for Real-world Tabular DataPreprint
Liu, Z., Liu, Q., Zhang, H., & Chen, Y. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2008.09775 · Cited by 17
“Most searching-based methods focus on generating second-order features [7], [13], [14], [15], [39], [40], [41].”
BackgroundMethodRandom Planted Forest: a directly interpretable tree ensemblePreprint
Hiabu, M., Mammen, E., & Meyer, J. T. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2012.14563 · Cited by 14
“Rpf differs from the tree based explainable boosting machine (Lou et al., 2012, 2013; Caruana et al., 2015; Lengerich et al., 2020) and the neural network based neural additive model (Agarwal et al., 2020) recently introduced in several ways.”
MethodExplainable AI and Adoption of Financial Algorithmic Advisors: an Experimental StudyPreprint
David, D. B., Resheff, Y. S., & Tron, T. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2101.02555 · Cited by 4
“learning, to develop fast and accurate techniques which explain black-box or opaque models [15, 20, 22], or train AI which is more interpretable by nature [1, 19].”
BackgroundHaas, S., Hegestweiler, K., Rapp, M. T., Muschalik, M., & Hüllermeier, E. (2024) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2024.1471208 · Cited by 6
Leveraging Real-World Evidence to Define Severe RSV Lower Respiratory Tract Disease in Adults
Panozzo, C. A., Walsh, E. E., Yang, Z., Wilson, E. M. P., Goswami, J., Stoszek, S., Loback, A., Ng, T., Francis, B. M., Simorellis, A. K., et al. (2024) · Infectious Diseases and Therapy · doi:10.1007/s40121-024-01072-4 · Cited by 3
“Logistic regression with ridge regularization and Explainable Boosting Machine [9] models were trained using various input variables (Supplemental Table 1).”
Mehta, B. Y., Yiyuan, Y., Pearce-Fisher, D., Ho, K., Goodman, S. M., Parks, M. L., Wang, F., Fontana, M. A., Ibrahim, S. A., Cram, P. M., et al. (2025) · Arthritis Care & Research · doi:10.1002/acr.25511 · Cited by 3
“46 49 We used EBMs because they have high accuracy with no loss of fi delity or approximations involved, are fully inter-pretable, and allow the importance of features to be measured both individually and aggregated as groups.”
Self-citationAdriko, K., Sedona, R., Seguini, L., Riedel, M., Cavallaro, G., & Paris, C. (2025) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2025.3624046 · Cited by 5
Validating Explainer Methods: A Functionally Grounded Approach for Numerical Forecasting
Haag, F., Hopf, K., & Staake, T. (2025) · Journal of Forecasting · doi:10.1002/for.70060 · Cited by 5
Kernbach, J. M., Schroeder, U., Hakvoort, K., Ort, J., Hamou, H. A., Bzdok, D., Temel, Y., Kubben, P., Weyland, C. S., Wiesmann, M., et al. (2025) · npj Precision Oncology · doi:10.1038/s41698-025-01183-2 · Cited by 4
“…Nu-support vector classi fi er), hyperplane learners (logistic regression), Gaussian Naive Bayes, a low-dimensional neural network, and tree-based algorithms with increasing model complexity (Decision Trees, random forest, AdaBoost, Gradient Boosting, and Explainable Boosting Machines) 31,62,63 .”
Dong, J., Xu-Wilson, M., Conroy, B., Khemani, R. G., & Newth, C. J. L. (2022) · Scientific Reports · doi:10.1038/s41598-022-13583-6 · Cited by 4
Çelik, M. F., Ișık, M. S., Erten, E., & Taşkın, G. (2023) · IEEE International Geoscience and Remote Sensing Symposium (IGARSS) · doi:10.1109/igarss52108.2023.10282371 · Cited by 3
“To address this limitation, [10] modified the equation by incorporating pairwise interactions between input features, resulting in Generalized Additive Models plus Interactions (GAM 2 s): In this equation, f i,j represents the pairwise interaction terms.”
MethodUnderstanding Interpretability in Modeling Biotechnological Processes: A Critical Review
Aguirre-Zapata, E., Álvarez, H., Lema-Pérez, L., di Sciascio, F. A., & Amicarelli, A. (2023) · Workshop on Information Processing and Control · doi:10.1109/rpic59053.2023.10530792 · Cited by 3
Fair and Actionable Causal Prescription Ruleset
Li, B., Levy, N., Youngmann, B., Galhotra, S., & Roy, S. (2025) · Proceedings of the ACM on Management of Data · doi:10.1145/3725293 · Cited by 6
How to improve PHEV electric mileage ratios? Factor decomposition with Explainable AI
Goto, H., Nishi, T., Shiga, T., Sasai, T., & Fukushima, S. (2025) · Transportation Research Part D Transport and Environment · doi:10.1016/j.trd.2025.104876 · Cited by 3
Musunuri, R., Zhu, B., Clarke, W. E., Hooper, W. F., Chu, T. R., Shelton, J. M., Corvelo, A., Chung, D., Sundar, S., Novak, A. M., et al. (2026) · NAR Genomics and Bioinformatics · doi:10.1093/nargab/lqag036 · Cited by 3
Less is Less: When are Snippets Insufficient for Human vs Machine Relevance Estimation?
Kazai, G., Mitra, B., Dong, A., Craswell, N., & Yang, L. (2022) · Lecture notes in computer science · doi:10.1007/978-3-030-99739-7_18 · Cited by 3
Why Data Scientists Prefer Glassbox Machine Learning
Caruana, R., & Nori, H. (2022) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3534678.3542627 · Cited by 3
Self-citationAn algorithm for training a class of polynomial models
Popescu, M. C., Grama, L., & Rusu, C. (2023) · Digital Signal Processing · doi:10.1016/j.dsp.2023.104168 · Cited by 3
A novel multi-featured decision system for multi-classification tasks
Xu, T. (2023) · Measurement Science and Technology · doi:10.1088/1361-6501/aceb11 · Cited by 3
Morlini, I., Albertson, S., & Orlandini, S. (2023) · Stochastic Environmental Research and Risk Assessment · doi:10.1007/s00477-023-02622-x · Cited by 3
“The routine ‘‘fitrgam’’ available in Matlab 2011b is used in the present study (Lou et al. 2012, 2013) and contributions to GAM predictions are determined by using the routine ‘‘plotlocaleffects.’’”
Oirbeek, R. V., Vandervorst, F., Bury, T., Willame, G., Grumiau, C., & Verdonck, T. (2024) · Risks · doi:10.3390/risks12050079 · Cited by 5
“Tree-based extended GAMs (GA 2 M) that detect interactions but often have limited loss function choices (Lou et al. 2013).”
Proxy endpoints — bridging clinical trials and real world data
Kryukov, M., Moriarty, K. P., Villamea, M., O'Dwyer, I., Chow, O., Dormont, F., Hernandez, R., Bar‐Joseph, Z., & Rufino, B. (2024) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2024.104723 · Cited by 3
Explaining the impact of design choices on model quality in predictive process monitoring
Kim, S., Comuzzi, M., & Francescomarino, C. D. (2024) · Journal of Intelligence and Information Systems · doi:10.1007/s10844-024-00903-7 · Cited by 5
“As a regression model, we consider both the eXtreme Gradient Boosting (XGB) and the Explainable Boosting Machine (EBM), a tree-based cyclic gradient boosting generalized additive model with automatic interaction detection (Lou et al., 2013).”
Comparative Analysis of Post Hoc Explainable Methods for Robotic Grasp Failure Prediction
Alvanpour, A., Acun, C., Spurlock, K., Robinson, C., Das, S. K., Popa, D. O., & Nasraoui, O. (2025) · Electronics · doi:10.3390/electronics14091868 · Cited by 5
Semantic Explanation for Deep Neural Networks Using Feature Interactions
Xia, B., Wang, X., & Yamasaki, T. (2021) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3474557 · Cited by 4
“Some works have tackled interpretation of complex models using features [26, 39].”
BackgroundZhang, X., Tan, S., Koch, P., Lou, Y., Chajewska, U., & Caruana, R. (2018) · arXiv (Cornell University) · Cited by 4
Self-citationDropout as a Regularizer of Interaction EffectsPreprint
Lengerich, B. J., Xing, E. P., & Caruana, R. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2007.00823 · Cited by 3
Self-citationLassi, A. (2023) · Revista Ciencias Sociales · doi:10.29166/csociales.v1i44.4213 · Cited by 5
Nguyen, M., Roman, G., & Soibam, B. (2023) · Behavioural Processes · doi:10.1016/j.beproc.2023.104944 · Cited by 2
Differentially private and explainable boosting machine with enhanced utility
Baek, I., & Chung, Y. D. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.128424 · Cited by 3
iTARGET: Interpretable Tailored Age Regression for Grouped Epigenetic Traits
Wu, Z., Herring, D. H., Spill, F., & Andrews, J. (2024) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm62325.2024.10821890 · Cited by 2
“We select different DNA CpG sites for each group to train the Explainable Boosting Machine (EBM) [21], [22], [23], improving prediction precision and enhancing interpretability and scalability.”
XAI-driven assessment of urban circular carbon economy: Using China's pilot cities as a case study
Wang, N., Zhai, H., Bai, Y., & Yu, F. (2025) · Urban Climate · doi:10.1016/j.uclim.2025.102407 · Cited by 2
Tuning hyperparameters and improving models
Maxwell, A. E., Ramezan, C. A., & He, Y. (2025) · Elsevier eBooks · doi:10.1016/b978-0-443-29306-1.00008-9 · Cited by 2
Discovering Phenotype-Specific Clinical Markers in Multiple Sclerosis
Giugliano, S., Basile, I., & Sannino, G. (2025) · IEEE Conference on Standards for Communications and Networking · doi:10.1109/cscn67557.2025.11230718 · Cited by 4
“…models such as Logistic Regression (LR) [22], Linear Discriminant Analysis (LDA) [23], DTree [24] and Explainable Boosting Machine (EBM) [25] were compared against black-box models including Multi-Layer Perceptron (MLP) [26], eXtreme Gradient Boosting (XGB) [27], Support Vector Machine…”
Explainable Artificial Intelligence for Deepfake Detection: Pipeline, Open Source and Comparisons
Li, H., Chen, J., Wang, B., García, S., & Camacho, D. (2026) · Expert Systems · doi:10.1111/exsy.70222 · Cited by 5
“GA2Ms that extend GAMs with automatically learned pairwise interactions trained using modern boosting strategies (Lou et al. 2013).”
Baucum, M., Rabiee, M., Aslani, B., & Dutta, K. (2026) · Decision Support Systems · doi:10.1016/j.dss.2026.114680 · Cited by 3
Interpretable Survival Gradient Boosting Models with Bagged Trees Base Learners
Jarmulski, W., & Wieczorkowska, A. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-48861-1_3 · Cited by 2
Preeclampsia Risk Prediction Using Machine Learning Methods Trained on Synthetic Data
Mazur-Milecka, M., Kowalczyk, N., Jaguszewska, K., Zamkowska, D., Wójcik, D. P., Preis, K., Skov, H. L., Wagner, S. R., Sandager, P., Sobotka, M., et al. (2023) · Lecture notes in networks and systems · doi:10.1007/978-3-031-38430-1_21 · Cited by 4
Relative Synergy Coefficient: A novel way to detect variable interaction in large dataset
Li, Y., Fu, K., Zhao, Y., & Yang, C. (2023) · Knowledge-Based Systems · doi:10.1016/j.knosys.2023.111112 · Cited by 2
Understanding the Impact of Design Choices on the Performance of Predictive Process Monitoring
Kim, S., Comuzzi, M., & Francescomarino, C. D. (2024) · Lecture notes in business information processing · doi:10.1007/978-3-031-56107-8_12 · Cited by 2
Impact Charts: A Tool for Identifying Systematic Bias in Social Systems and Data
Vengroff, D. E. (2024) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3630106.3658965 · Cited by 3
“A related approach is to use bagging with carefully designed weak but inherently explainable models [34, 35] rather than SHAP on a more general set of modeling techniques.”
MethodGlass box machine learning and corporate bond returns
Bell, S., Kakhbod, A., Lettau, M., & Nazemi, A. (2026) · Journal of Financial Economics · doi:10.1016/j.jfineco.2026.104294 · Cited by 10
Ontologies in modelling and analysing of big genetic data
Podkolodnyy, N. L., Podkolodnaya, O. A., Иванисенко, В. А., & Марченко, М. А. (2025) · Vavilov Journal of Genetics and Breeding · doi:10.18699/vjgb-24-101 · Cited by 4
Robust Feature Selection for Classifying Early Ischemic Changes in Posterior Stroke
Rist, L., Ditt, H., Sühling, M., Honus, V., Schramm, P., Maier, A., & Taubmann, O. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-81101-2_7 · Cited by 2
Exploring explanation deficits in subclinical mastitis detection with explainable boosting machines
Jin, C., Upton, J., & Namee, B. M. (2025) · Discover Agriculture · doi:10.1007/s44279-025-00246-z · Cited by 2
“Models are trained using the Explainable Boosting Machine (EBM) algorithm [15, 20], a state of the art explainable machine learning method, and the explanations that can be generated from models trained using different feature sets is examined.”
Qualifying Training Datasets for Data-Driven Turbulence Closures
Banerjee, T., Ray, J., Barone, M., & Domino, S. P. (2022) · AIAA AVIATION Forum · doi:10.2514/6.2022-3979 · Cited by 2
DPPred: An Effective Prediction Framework with Concise Discriminative Patterns
Shang, J., Jiang, M., Tong, W., Xiao, J., Peng, J., & Han, J. (2017) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2017.2757476 · Cited by 5
From Black-Box to White-Box: Interpretable Learning with Kernel Machines
Zhang, H., Nakadai, S., & Fukumizu, K. (2018) · Lecture notes in computer science · doi:10.1007/978-3-319-96136-1_18 · Cited by 3
A Scalable Algorithm for Higher-order Features Generation using MinHash
Pooja, A., Nair, N., & Rastogi, R. R. (2018) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3269206.3271752 · Cited by 2
“By analyzing the running time and memory requirements, we established that the proposed scheme leads to a significant reduction in time and memory requirement compared to alternatives like FAST, Random Forest, Deep Crossing model and exhaustive search.”
BackgroundMethodInfluentialSymbolic Metamodels for Interpreting Black-Boxes Using Primitive Functions
Abroshan, M., Mishra, S., & Khalili, M. M. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i6.25816 · Cited by 4
Post-Hoc Interpretation of Transformer Hyperparameters with Explainable Boosting Machines
Deb, K., Zhang, X., & Duh, K. (2022) · BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP (BlackboxNLP) · doi:10.18653/v1/2022.blackboxnlp-1.5 · Cited by 2
“We use EBM as introduced in (Lou et al., 2012, 2013; Caruana et al., 2015) and implemented in Nori et al. (2019).”
MethodAutomated and intelligent system for World Health Organization data forecasting
Soares, F., & de Freitas, H. C. (2023) · Expert Systems · doi:10.1111/exsy.13521 · Cited by 2
“As a result of the work, the authors applied the method GA 2 M (generalized additive models plus interactions), proposed in Lou et al. (2013) and obtained good results in predictions for individual patients in a concise way.”
MethodIntrinsically Interpretable Models for Credit Risk Assessment
Duvnjak, M., Merćep, A., & Kostanjčar, Z. (2024) · International Convention on Information and Communication Technology, Electronics and Microelectronics · doi:10.1109/mipro60963.2024.10569726 · Cited by 4
“Contri-butions made by [27]: built an interpretable model with high predictive power.”
BackgroundMethodInfluentialFocusLearn: Fully-Interpretable, High-Performance Modular Neural Networks for Time Series
Su, Q.-Q., Kloukinas, C., & Garcez, A. S. D. (2023) · IEEE International Joint Conference on Neural Network · doi:10.1109/ijcnn60899.2024.10651481 · Cited by 10
“GAM variants such as NAM [11], GA 2 M [26], Neural Interaction Transparency [27] and Neural Basis Model [28], are computationally efficient and can handle large data sets.”
Improving Competence via Iterative State Space Refinement
Basich, C., Svegliato, J., Beach, A., Wray, K. H., Witwicki, S., & Zilberstein, S. (2021) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros51168.2021.9636239 · Cited by 3
“We implemented our feedback profile λ as a GA2M [15], which works well on low dimensional feature spaces where pairwise feature interactions are important.”
MethodLogic Constraints to Feature Importance
Picchiotti, N., & Gori, M. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-08421-8_27 · Cited by 2
Pairwise interactions origin of entropy functionsPreprint
PYKH, Y. A. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1506.05731 · Cited by 3
“actions. Social interactions models. Zhao et al. in [39] consider a general model of pairwise human social interaction consisting of N agents representing interacting in social network. Lou et al. in [40] analyze standard generalized additive models (GAMs) with pairwise interactions. As we have briefly discussed above…”
BackgroundTree Boosted Varying Coefficient ModelsPreprint
Zhou, Y., & Hooker, G. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1904.01058 · Cited by 2
“There is a sizable literature on this topic, from the appearance of local methods (Ribeiro et al., 2016) to recent applications on neural nets (Zhang and Zhu, 2018), random forests (Mentch and Hooker, 2016; Basu et al., 2018) and complex model distillation (Lou et al., 2012, 2013; Tan et al., 2017).”
BackgroundMethodSelf-citationExplainable AI and Adoption of Algorithmic Advisors: an Experimental Study.Preprint
David, D. B., Resheff, Y. S., & Tron, T. (2021) · arXiv (Cornell University) · Cited by 2
Distilling Interpretable Models into Human-Readable CodePreprint
Ravina, W., Sterling, E., Oryeshko, O., Bell, N., Zhuang, H., Wang, X., Wu, Y., & Grushetsky, A. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2101.08393 · Cited by 3
“Both FICO and MSLR-WEB30K have been previously studied in the context of interpretability [2, 7, 18, 31].”
BackgroundInterpretable Machine Learning: Moving From Mythos to DiagnosticsPreprint
Chen, V., Li, J., Kim, J. S., Plumb, G., & Talwalkar, A. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2103.06254 · Cited by 15
“Another method, Generalized additive model (GAM), and its variant (GA2M) (Lou et al., 2013) benefit from being able to represent a prediction in terms of univariate features and pairwise interactions.”
Partially Interpretable Estimators (PIE): Black-Box-Refined Interpretable Machine LearningPreprint
Wang, T., Yang, J., Li, Y., & Wang, B. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2105.02410 · Cited by 7
“We will compare with a classic GAM, an improved version of GA2M, called Explainable Boosted Machine (EBM) and the most recent NAM in this paper.”
BackgroundInfluentialVetter, J., Lim, K., Dijkstra, T. M. H., Dargaville, P. A., Kohlbacher, O., Macke, J. H., & Poets, C. F. (2023) · medRxiv · doi:10.1101/2023.03.14.23287021 · Cited by 2
Costa, M., Siebert, F. W., Azevedo, C. L., Marques, M., & Moura, F. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4886041 · Cited by 2
Early prediction of ovarian cancer risk based on real world dataPreprint
de la Oliva, V., Esteban-Medina, A., Alejos, L., Muñoyerro-Muñiz, D., Villegas, R., Dopazo, J., & Loucera, C. (2024) · medRxiv · doi:10.1101/2024.07.26.24310994 · Cited by 4
“An Explainable Boosting Machine (EBM) algorithm [27] was used for its capacity to provide transparent explanations of its predictive decisions.”
MethodA Novel Test for Additivity in Supervised Ensemble LearnersPreprint
Mentch, L., & Hooker, G. (2014) · arXiv (Cornell University) · Cited by 2
Self-citationUnderstanding Learned Models by Identifying Important Features at the Right Resolution
Lee, K., Sood, A., & Craven, M. W. (2019) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v33i01.33014155 · Cited by 5
IFME: Influence Function Based Model Explanation for Black Box Decision Systems
Zha, B., & Shen, H. (2020) · Communications in computer and information science · doi:10.1007/978-981-15-2767-8_27 · Cited by 2
A simulation-based tree method for building linear models with interactions
Wang, J., Cabrera, J., & Tsui, K. (2020) · Communication in Statistics- Theory and Methods · doi:10.1080/03610926.2020.1749665 · Cited by 1
Geometrically interpretable Variance Hyper Rectangle learning for pattern classification
Sun, J., Gu, H., Peng, H., Fang, Y., & Wang, X. (2022) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2022.105494 · Cited by 2
Chang, T. L., Xia, H., Mahajan, S., Mahajan, R., Maisog, J., Vattikuti, S., Chow, C. C., & Chang, J. C. (2024) · PLoS ONE · doi:10.1371/journal.pone.0302871 · Cited by 5
Cilínio, M., Pereira, M. A. G., Duarte, D., Mata, L. A., & Vieira, P. A. (2024) · AEU - International Journal of Electronics and Communications · doi:10.1016/j.aeue.2024.155339 · Cited by 1
Metadata
- Authors
- Lou, Y., Caruana, R., Gehrke, J., & Hooker, G.
- Affiliations
- Cornell University (US)
- Microsoft (United States) (US)
- Venue
- Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining
- Publisher
- ACM
- Published
- 2013-08-11
- Pages
- 623-631
- OpenAlex
- W2046945713
- Cited by
- 564
- References
- 22
- Citations by year
- 2026: 66 · 2025: 78 · 2024: 88 · 2023: 87 · 2022: 65 · 2021: 57 · 2020: 45 · 2019: 29 · 2018: 24 · 2017: 11 · 2016: 8 · 2015: 4 · 2014: 2
- FWCI
- 3.01
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Applications
- Physical Sciences › Computer Science › Computer Graphics and Computer-Aided Design › Data Visualization and Analytics
- Keywords
- generalized additive models, pairwise interactions, model interpretability, model accuracy
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.
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.