{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,24]],"date-time":"2026-01-24T18:32:10Z","timestamp":1769279530212,"version":"3.49.0"},"reference-count":33,"publisher":"Springer Science and Business Media LLC","issue":"12","license":[{"start":{"date-parts":[[2023,4,25]],"date-time":"2023-04-25T00:00:00Z","timestamp":1682380800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,4,25]],"date-time":"2023-04-25T00:00:00Z","timestamp":1682380800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Soft Comput"],"published-print":{"date-parts":[[2023,6]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Uncertainty estimation has received momentous consideration in applied machine learning to capture model uncertainty. For instance, the Monte-Carlo dropout method (MC-dropout), an approximated Bayesian approach, has gained intensive attention in producing model uncertainty due to its simplicity and efficiency. However, MC-dropout has revealed shortcomings in capturing erroneous predictions lying in the overlapping classes. Such predictions underlie noisy data points that can neither be reduced by more training data nor detected by model uncertainty. On the other hand, Monte-Carlo based on adversarial attacks (MC-AA), an outstanding method, performs perturbations on the inputs using the adversarial attack idea to capture model uncertainty. This method admittedly mitigates the shortcomings of the previous methods by capturing wrong labels in overlapping regions. Motivated by this method that was only validated with neural networks, we sought to apply MC-AA on various graph neural network models to obtain uncertainties using two public real-world graph datasets known as Elliptic and GitHub. First, we perform binary node classifications, then we apply MC-AA and other recent uncertainty estimation methods to capture the uncertainty of the models. Uncertainty evaluation metrics are computed to evaluate and compare the performance of the uncertainty of the model. We highlight the efficacy of MC-AA in capturing uncertainties in graph neural networks wherein MC-AA outperforms other given methods.<\/jats:p>","DOI":"10.1007\/s00500-023-08031-0","type":"journal-article","created":{"date-parts":[[2023,4,25]],"date-time":"2023-04-25T14:04:34Z","timestamp":1682431474000},"page":"7925-7937","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":3,"title":["Uncertainty estimation-based adversarial attacks: a viable approach for graph neural networks"],"prefix":"10.1007","volume":"27","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-8320-6423","authenticated-orcid":false,"given":"Ismail","family":"Alarab","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Simant","family":"Prakoonwit","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2023,4,25]]},"reference":[{"key":"8031_CR1","doi-asserted-by":"crossref","unstructured":"Abdar M, Pourpanah F, Hussain S, Rezazadegan D, Liu L, Ghavamzadeh M, Fieguth P, Cao X, Khosravi A, Acharya UR et al. (2021) A review of uncertainty quantification in deep learning: techniques, applications and challenges, Information Fusion","DOI":"10.1016\/j.inffus.2021.05.008"},{"issue":"2","key":"8031_CR2","doi-asserted-by":"publisher","first-page":"1001","DOI":"10.1007\/s11063-021-10424-x","volume":"53","author":"I Alarab","year":"2021","unstructured":"Alarab I, Prakoonwit S, Nacer MI (2021) Illustrative discussion of mc-dropout in general dataset: uncertainty estimation in bitcoin. Neural Process Lett 53(2):1001\u20131011","journal-title":"Neural Process Lett"},{"key":"8031_CR3","doi-asserted-by":"crossref","unstructured":"Alarab I, Prakoonwit S (2021) Adversarial attack for uncertainty estimation: Identifying critical regions in neural networks. arXiv:2107.07618","DOI":"10.1007\/s11063-021-10707-3"},{"key":"8031_CR4","doi-asserted-by":"crossref","unstructured":"Alarab I, Prakoonwit S, Nacer MI (2020a) Competence of graph convolutional networks for anti-money laundering in bitcoin blockchain, In: Proceedings of the 2020a 5th international conference on machine learning technologies, pp 23\u201327","DOI":"10.1145\/3409073.3409080"},{"key":"8031_CR5","doi-asserted-by":"crossref","unstructured":"Alarab I, Prakoonwit S, Nacer MI (2020b) Comparative analysis using super- vised learning methods for anti-money laundering in bitcoin, In: Proceedings of the 2020b 5th international conference on machine learning technologies, pp 11\u201317","DOI":"10.1145\/3409073.3409078"},{"key":"8031_CR6","unstructured":"Chakraborty A, Alam M, Dey V, Chattopadhyay A, Mukhopadhyay D (2018) Adversarial attacks and defences: a survey, arXiv preprint arXiv:1810.00069"},{"key":"8031_CR7","unstructured":"Du J, Zhang S, Wu G, Moura JMF, Kar S (2018) Topology adaptive graph convolutional networks. arXiv:1710.10370"},{"key":"8031_CR8","unstructured":"Fey M, Lenssen JE (2019) Fast graph representation learning with PyTorch Geometric, In: ICLR workshop on representation learning on graphs and manifolds"},{"issue":"3","key":"8031_CR9","first-page":"4","volume":"1","author":"Y Gal","year":"2016","unstructured":"Gal Y (2016) Uncertainty in deep learning. Univ Camb 1(3):4","journal-title":"Univ Camb"},{"key":"8031_CR10","unstructured":"Gal Y, Ghahramani Z (2016) Dropout as a Bayesian approximation: Representing model uncertainty in deep learning, in: international conference on machine learning, PMLR, pp 1050\u20131059"},{"key":"8031_CR11","unstructured":"Gal Y, Hron J, Kendall A (2017) Concrete dropout, arXiv preprint arXiv:1705.07832"},{"key":"8031_CR12","unstructured":"Github social network. URL https:\/\/snap.stanford.edu\/data\/github-social.html"},{"issue":"4","key":"8031_CR13","doi-asserted-by":"publisher","first-page":"2776","DOI":"10.1109\/TII.2020.2986316","volume":"17","author":"Z Guo","year":"2020","unstructured":"Guo Z, Wang H (2020) A deep graph neural network-based mechanism for social recommendations. IEEE Trans Ind Inf 17(4):2776\u20132783","journal-title":"IEEE Trans Ind Inf"},{"key":"8031_CR14","unstructured":"Hamilton WL, Ying R, Leskovec J (2018) Inductive representation learning on large graphs. arXiv:1706.02216"},{"issue":"3","key":"8031_CR15","first-page":"1","volume":"21","author":"S Harada","year":"2020","unstructured":"Harada S, Akita H, Tsubaki M, Baba Y, Takigawa I, Yamanishi Y, Kashima H (2020) Dual graph convolutional neural network for predicting chemical networks. BMC Bioinform 21(3):1\u201313","journal-title":"BMC Bioinform"},{"key":"8031_CR16","unstructured":"Kennamer N, Ihler AT, Kirkby D (2019) Empirical study of mc-dropout in various astronomical observing conditions. In: CVPR workshops, pp 17\u201320"},{"key":"8031_CR17","unstructured":"Kipf TN, Welling M (2017) Semi-supervised classification with graph convolu- tional networks. arXiv:1609.02907"},{"issue":"9","key":"8031_CR18","doi-asserted-by":"publisher","first-page":"3981","DOI":"10.1021\/acs.jcim.9b00387","volume":"59","author":"J Lim","year":"2019","unstructured":"Lim J, Ryu S, Park K, Choe YJ, Ham J, Kim WY (2019) Predicting drug\u2014target interaction using a novel graph neural network with 3d structure embedded graph representation. J Chem Inf Model 59(9):3981\u20133988","journal-title":"J Chem Inf Model"},{"issue":"3","key":"8031_CR19","doi-asserted-by":"publisher","first-page":"448","DOI":"10.1162\/neco.1992.4.3.448","volume":"4","author":"DJ MacKay","year":"1992","unstructured":"MacKay DJ (1992) A practical bayesian framework for backpropagation net-works. Neural Comput 4(3):448\u2013472","journal-title":"Neural Comput"},{"key":"8031_CR20","unstructured":"Mobiny A, Nguyen HV, Moulik S, Garg N, Wu CC (2019) Dropconnect is effective in modeling uncertainty of Bayesian deep networks, arXiv preprint arXiv:1906.04569"},{"key":"8031_CR21","doi-asserted-by":"crossref","unstructured":"Morris C, Ritzert M, Fey M, Hamilton WL, Lenssen JE, Rattan G, Grohe M (2020) Weisfeiler and leman go neural: higher-order graph neural networks. arXiv:1810.02244","DOI":"10.1609\/aaai.v33i01.33014602"},{"key":"8031_CR22","volume-title":"Bayesian learning for neural networks","author":"RM Neal","year":"2012","unstructured":"Neal RM (2012) Bayesian learning for neural networks, vol 118. Springer Sci-ence & Business Media, Berlin"},{"key":"8031_CR23","unstructured":"Ng M, Guo F, Biswas L, Wright GA (2018) Estimating uncertainty in neu- ralnetworks for segmentation quality control, Tech. rep., Technical report"},{"key":"8031_CR24","unstructured":"Paszke A, Gross S, Massa F, Lerer A, Bradbury J, Chanan G, Killeen T, Lin Z, Gimelshein N, Antiga L, Desmaison A, Kopf A, Yang E, DeVito Z, Raison M, Tejani A, Chilamkurthy S, Steiner B, Fang L, Bai J, Chintala S (2019) Pytorch: an imperative style, high- performance deep learning library, In: Wallach H, Larochelle H, Beygelzimer A, d'Alch\u00b4e-Buc F, Fox E, Garnett R (Eds.), Advances in Neural Information Processing Systems 32, Curran Associates, Inc., 2019, pp 8024\u20138035. http:\/\/papers.neurips.cc\/paper\/9015-pytorch-an-imperative-style-high-performance-deep-learning-library.pdf"},{"key":"8031_CR25","doi-asserted-by":"crossref","unstructured":"Ranjan E, Sanyal S, Talukdar PP (2020) Asap: adaptive structure aware pooling for learning hierarchical graph representations. arXiv:1911.07979","DOI":"10.1609\/aaai.v34i04.5997"},{"key":"8031_CR26","doi-asserted-by":"crossref","unstructured":"Rasmussen CE (2003) Gaussian processes in machine learning, In: Summer school on machine learning, Springer, pp 63\u201371","DOI":"10.1007\/978-3-540-28650-9_4"},{"key":"8031_CR27","unstructured":"Rozemberczki B, Allen C, Sarkar R (2019) Multi-scale attributed node embed- ding. arXiv:1909.13021"},{"key":"8031_CR28","unstructured":"Smith L, Gal Y (2018) Understanding measures of uncertainty for adversarial example detection, arXiv preprint arXiv:1803.08533"},{"key":"8031_CR29","unstructured":"Van Amersfoort J, Smith L, Teh YW, Gal Y (2020) Uncertainty estimation using a single deep deterministic neural network, In: International conference on machine learning, PMLR, pp 9690\u20139700"},{"key":"8031_CR30","unstructured":"van Amersfoort J, Smith L, Jesson A, Key O, Gal Y (2021) Improving de- terministic uncertainty estimation in deep learning for classification and regression, arXiv preprint arXiv:2102.11409"},{"key":"8031_CR31","unstructured":"Vlickovic P, Cucurull G, Casanova A, Romero A, Li`o P, Bengio Y (2018) Graph attention networks. arXiv:1710.10903"},{"key":"8031_CR32","unstructured":"Weber M, Domeniconi G, Chen J, Weidele DKI, Bellei C, Robin-son T, Leiserson CE (2019) Anti-money laundering in bitcoin: Experimenting with graph convolutional networks for financial forensics, arXiv preprint arXiv:1908.02591"},{"key":"8031_CR33","unstructured":"Xu K, Hu W, Leskovec J, Jegelka S (2018) How powerful are graph neural networks?, arXiv preprint arXiv:1810.00826"}],"container-title":["Soft Computing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00500-023-08031-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00500-023-08031-0\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00500-023-08031-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,5,17]],"date-time":"2023-05-17T14:30:06Z","timestamp":1684333806000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00500-023-08031-0"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,4,25]]},"references-count":33,"journal-issue":{"issue":"12","published-print":{"date-parts":[[2023,6]]}},"alternative-id":["8031"],"URL":"https:\/\/doi.org\/10.1007\/s00500-023-08031-0","relation":{},"ISSN":["1432-7643","1433-7479"],"issn-type":[{"value":"1432-7643","type":"print"},{"value":"1433-7479","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,4,25]]},"assertion":[{"value":"10 March 2023","order":1,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"25 April 2023","order":2,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare that they have no conflict of interest.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}},{"value":"This paper does not contain any studies with human participants or animals performed by any of the authors.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethical approval"}},{"value":"Informed consent was obtained from all individual participants included in the study.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Informed consent"}}]}}