The DOI matches its Crossref record; 35 of the 37 references it cites were found in a registry or an index.
Enhancing Explainability of Neural Networks Through Architecture Constraints
Open accessAbstract
Prediction accuracy and model explainability are the two most important objectives when developing machine learning algorithms to solve real-world problems. Neural networks are known to possess good prediction performance but suffer from a lack of model interpretability. In this article, we propose to enhance the explainability of neural networks through the following architecture constraints: 1) sparse additive subnetworks; 2) projection pursuit with orthogonality constraint; and 3) smooth function approximation. It leads to an enhanced explainable neural network (ExNN) with a superior balance between prediction performance and model interpretability. We derive sufficient identifiability conditions for the proposed ExNN model. The multiple parameters are simultaneously estimated by a modified minibatch gradient descent method based on the backpropagation algorithm for calculating the derivatives and the Cayley transform for preserving the projection orthogonality. Through simulation study under six different scenarios, we compare the proposed method to several benchmarks, including least absolute shrinkage and selection operator, support vector machine, random forest, extreme learning machine, and multilayer perceptron. It is shown that the proposed ExNN model keeps the flexibility of pursuing high prediction accuracy while attaining improved interpretability. Finally, a real data example is employed as a showcase application.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“However, in tightly controlled domains like IoT systems, avionics, finance, banking, autonomous systems, and medical services, integrating AI and its interoperability with legacy applications is much more onerous [27], [47].”
From A Review of Trustworthy and Explainable Artificial Intelligence (XAI) (2023) · cites this paper as [47]
Experimental — accuracy not yet measuredThe check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“Leading to significant biases and prejudices in the training set, black-box models may still make poor predictions even in the absence of adversarial attacks [28], [46], [47].”
From A Review of Trustworthy and Explainable Artificial Intelligence (XAI) (2023) · cites this paper as [47]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
Greedy function approximation: A gradient boosting machine
Friedman, J. H. (2001) · The Annals of Statistics · doi:10.1214/aos/1013203451 · Cited by 30,731
The Elements of Statistical Learning
Hastie, T., Tibshirani, R., & Friedman, J. (2009) · Springer Series in Statistics · doi:10.1007/978-0-387-84858-7 · Cited by 25,970
"Why Should I Trust You?": Explaining the Predictions of Any Classifier
Ribeiro, M. T., Singh, S., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939778 · Cited by 17,555
Distilling the Knowledge in a Neural Network
Hinton, G., Vinyals, O., & Dean, J. (2015) · arXiv · doi:10.48550/arxiv.1503.02531 · Cited by 14,204
(2010) · Cited by 14,134
Extreme learning machine: Theory and applications
Huang, G.-B., Zhu, Q.-Y., & Siew, C.-K. (2006) · Neurocomputing · doi:10.1016/j.neucom.2005.12.126 · Cited by 13,285
The Elements of Statistical Learning: Data Mining, Inference, and Prediction
Lu, Z. Q. J. (2010) · Journal of the Royal Statistical Society Series A (Statistics in Society) · doi:10.1111/j.1467-985x.2010.00646_6.x · Cited by 12,916
Brown, R. A., Hastie, T. J., & Tibshirani, R. J. (1991) · Biometrics · doi:10.2307/2532174 · Cited by 8,263
Koyak, R. A., Hastie, T. J., & Tibshirani, R. J. (1991) · Journal of the American Statistical Association · doi:10.2307/2290538 · Cited by 7,717
Approximation capabilities of multilayer feedforward networks
Hornik, K. (1991) · Neural Networks · doi:10.1016/0893-6080(91)90009-t · Cited by 6,091
Spline Models for Observational Data.
Müller, H., & Wahba, G. (1991) · Journal of the American Statistical Association · doi:10.2307/2290434 · Cited by 4,992
Spline Models for Observational Data
Wahba, G. (1990) · doi:10.1137/1.9781611970128 · Cited by 4,184
Sparse Principal Component Analysis
Zou, H., Hastie, T., & Tibshirani, R. (2006) · Journal of Computational and Graphical Statistics · doi:10.1198/106186006x113430 · Cited by 3,152
Friedman, J. H., & Stuetzle, W. (1981) · Journal of the American Statistical Association · doi:10.1080/01621459.1981.10477729 · Cited by 2,116
Buciluǎ, C., Caruana, R., & Niculescu-Mizil, A. (2006) · Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining · doi:10.1145/1150402.1150464 · Cited by 2,068
Goldstein, A., Kapelner, A., Bleich, J., & Pitkin, E. (2015) · Journal of Computational and Graphical Statistics · doi:10.1080/10618600.2014.907095 · Cited by 1,861
Do Deep Nets Really Need to be Deep?Preprint
Ba, J. L., & Caruana, R. (2013) · arXiv (Cornell University) · doi:10.48550/arxiv.1312.6184 · Cited by 1,592
Visualizing the effects of predictor variables in black box supervised learning models
Apley, D. W., & Zhu, J. (2020) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/rssb.12377 · Cited by 1,535
DARPA's Explainable Artificial Intelligence Program
Gunning, D., & Aha, D. W. (2019) · AI Magazine · doi:10.1609/aimag.v40i2.2850 · Cited by 1,360
Semiparametric least squares (SLS) and weighted SLS estimation of single-index models
Ichimura, H. (1993) · Journal of Econometrics · doi:10.1016/0304-4076(93)90114-k · Cited by 1,175
A feasible method for optimization with orthogonality constraints
Wen, Z., & Yin, W. (2012) · Mathematical Programming · doi:10.1007/s10107-012-0584-1 · Cited by 897
A Note on the Use of Principal Components in Regression
Jolliffe, I. T. (1982) · Journal of the Royal Statistical Society Series C (Applied Statistics) · doi:10.2307/2348005 · Cited by 856
Prediction by Supervised Principal Components
Bair, E., Hastie, T. J., Paul, D., & Tibshirani, R. J. (2006) · Journal of the American Statistical Association · doi:10.1198/016214505000000628 · Cited by 824
Ravikumar, P., Lafferty, J., Liu, H., & Wasserman, L. (2009) · Journal of the Royal Statistical Society Series B (Statistical Methodology) · doi:10.1111/j.1467-9868.2009.00718.x · Cited by 548
Barshan, E., Ghodsi, A., Azimifar, Z., & Jahromi, M. Z. (2010) · Pattern Recognition · doi:10.1016/j.patcog.2010.12.015 · Cited by 343
Regression modeling in back-propagation and projection pursuit learning
Hwang, J., Lay, S.-R., Maechler, M., Martin, R. D., & Schimert, J. (1994) · IEEE Transactions on Neural Networks · doi:10.1109/72.286906 · Cited by 282
Visualizing the Effects of Predictor Variables in Black Box Supervised Learning ModelsPreprint
Apley, D. W., & Zhu, J. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1612.08468 · Cited by 247
A Novel Pruning Algorithm for Smoothing Feedforward Neural Networks Based on Group Lasso Method
Wang, J., Xu, C., Xi-feng, Y., & Żurada, J. M. (2017) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2017.2748585 · Cited by 99
Explainable Neural Networks based on Additive Index ModelsPreprint
Vaughan, J., Sudjianto, A., Brahimi, E., Chen, J., & Nair, V. N. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1806.01933 · Cited by 74
Self-citationFernández‐Navarro, F., Hervás‐Martínez, C., Sánchez‐Monedero, J., & Gutiérrez, P. A. (2011) · Neurocomputing · doi:10.1016/j.neucom.2010.11.032 · Cited by 73
Generalized Additive and Index Models with Shape Constraints
Chen, Y., & Samworth, R. J. (2015) · Journal of the Royal Statistical Society Series B (Statistical Methodology) · doi:10.1111/rssb.12137 · Cited by 71
Locally Interpretable Models and Effects based on Supervised Partitioning (LIME-SUP)Preprint
Hu, L., Chen, J., Nair, V. N., & Sudjianto, A. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1806.00663 · Cited by 38
Self-citationOn the identifiability of additive index models
Yuan, M. (2011) · Statistica Sinica · doi:10.5705/ss.2008.117 · Cited by 33
Liu, X., Chen, J., Vaughan, J., Nair, V. N., & Sudjianto, A. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1808.07216 · Cited by 5
Self-citation
Research works printed but not linked: 1
- brainard, 2018, What Are We Learning About Artificial Intelligence in Financial Services?
Cited by
Sorted by citations
Explainable artificial intelligence: a comprehensive review
Minh, D., Wang, H. X., Li, Y. F., & Nguyen, T. N. (2022) · Artificial Intelligence Review · doi:10.1007/s10462-021-10088-y · Cited by 983
Kufel, J., Bargieł-Łączek, K., Kocot, S., Koźlik, M., Bartnikowska, W., Janik, M., Czogalik, Ł., Dudek, P., Magiera, M., Lis, A., et al. (2023) · Diagnostics · doi:10.3390/diagnostics13152582 · Cited by 476
“Deep neural networks, particularly with multiple hidden layers, which can be considered as black boxes, lack interpretability [86,87].”
BackgroundA Review of Trustworthy and Explainable Artificial Intelligence (XAI)
Chamola, V., Hassija, V., Sulthana, A. R., Ghosh, D., Dhingra, D., & Sikdar, B. (2023) · IEEE Access · doi:10.1109/access.2023.3294569 · Cited by 352
“For instance, the European Union’s GDPR framework, which was been passed in 2016 and went into effect in May 2018, established criteria to support the ‘‘entitlement of an explanation’’ premise for users [47].”
BackgroundMethodIslam, M. R., Ahmed, M. U., Barua, S., & Begum, S. (2022) · Applied Sciences · doi:10.3390/app12031353 · Cited by 366
Degas, A., Islam, M. R., Hurter, C., Barua, S., Rahman, H., Poudel, M., Ruscio, D., Ahmed, M. U., Begum, S., Rahman, M. A., et al. (2022) · Applied Sciences · doi:10.3390/app12031295 · Cited by 168
Metadata
- Authors
- Yang, Z., Zhang, A., & Sudjianto, A.
- Affiliations
- University of Hong Kong (HK)
- Wells Fargo (United States) (US)
- Venue
- IEEE Transactions on Neural Networks and Learning Systems
- Published
- 2021-06
- Volume
- 32
- Issue
- 6
- Pages
- 2610-2621
- ISSN
- 2162-237X (ISSN-L)
- 2162-2388
- OpenAlex
- W3045825034
- Cited by
- 106
- References
- 37
- Citations by year
- 2026: 13 · 2025: 6 · 2024: 19 · 2023: 22 · 2022: 15 · 2021: 22
- FWCI
- 6.68
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Transactions on Neural Networks and Learning Systems (publishedVersion)
- PubMed (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Big Data Project Fund of The University of Hong Kong through Dr. Patrick Poon’s Donation
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Stochastic Gradient Optimization Techniques
- Keywords
- explainable neural networks, projection pursuit, orthogonality constraints, smooth function approximation, identifiability conditions, Cayley transform, backpropagation, model interpretability, prediction 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
