{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,28]],"date-time":"2026-09-28T23:21:54Z","timestamp":1790637714401,"version":"4.1.0"},"reference-count":58,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2020,1,1]],"date-time":"2020-01-01T00:00:00Z","timestamp":1577836800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2020,1,1]],"date-time":"2020-01-01T00:00:00Z","timestamp":1577836800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2020,1,1]],"date-time":"2020-01-01T00:00:00Z","timestamp":1577836800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100012166","name":"National Basic Research Program of China","doi-asserted-by":"publisher","award":["2017YFA0700904"],"award-info":[{"award-number":["2017YFA0700904"]}],"id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["61836014"],"award-info":[{"award-number":["61836014"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["61621136008"],"award-info":[{"award-number":["61621136008"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["61620106010"],"award-info":[{"award-number":["61620106010"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Beijing Academy of Artificial Intelligence"},{"DOI":"10.13039\/100004318","name":"Microsoft Research Asia","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100004318","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. on Image Process."],"published-print":{"date-parts":[[2020]]},"DOI":"10.1109\/tip.2020.2993098","type":"journal-article","created":{"date-parts":[[2020,5,13]],"date-time":"2020-05-13T22:37:01Z","timestamp":1589409421000},"page":"6707-6720","source":"Crossref","is-referenced-by-count":15,"title":["Interpret Neural Networks by Extracting Critical Subnetworks"],"prefix":"10.1109","volume":"29","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-5516-7284","authenticated-orcid":false,"given":"Yulong","family":"Wang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hang","family":"Su","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bo","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4907-7354","authenticated-orcid":false,"given":"Xiaolin","family":"Hu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref39","first-page":"91","article-title":"Faster R-CNN: Towards real-time object detection with region proposal networks","author":"ren","year":"2015","journal-title":"Proc NIPS"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.249"},{"key":"ref33","article-title":"UMAP: Uniform manifold approximation and projection for dimension reduction","author":"mcinnes","year":"2018","journal-title":"arXiv 1802 03426"},{"key":"ref32","article-title":"Towards deep learning models resistant to adversarial attacks","author":"madry","year":"2017","journal-title":"arXiv 1706 06083"},{"key":"ref31","first-page":"2579","article-title":"Visualizing data using t-SNE","volume":"9","author":"ven der maaten","year":"2008","journal-title":"J Mach Learn Res"},{"key":"ref30","article-title":"Characterizing adversarial subspaces using local intrinsic dimensionality","author":"ma","year":"2018","journal-title":"arXiv 1801 02613"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.282"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1016\/j.dsp.2017.10.011"},{"key":"ref35","doi-asserted-by":"crossref","first-page":"529","DOI":"10.1038\/nature14236","article-title":"Human-level control through deep reinforcement learning","volume":"518","author":"mnih","year":"2015","journal-title":"Nature"},{"key":"ref34","article-title":"On detecting adversarial perturbations","author":"metzen","year":"2017","journal-title":"arXiv 1702 04267"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.298"},{"key":"ref27","article-title":"Pruning filters for efficient ConvNets","author":"li","year":"2016","journal-title":"arXiv 1608 08710"},{"key":"ref29","article-title":"An unexpected unity among methods for interpreting model predictions","author":"lundberg","year":"2016","journal-title":"arXiv 1611 07478"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.354"},{"key":"ref1","first-page":"1","article-title":"Towards better understanding of gradient-based attribution methods for deep neural networks","author":"ancona","year":"2018","journal-title":"Proc ICLR"},{"key":"ref20","article-title":"Adam: A method for stochastic optimization","author":"kingma","year":"2014","journal-title":"arXiv 1412 6980"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.172"},{"key":"ref21","first-page":"1885","article-title":"Understanding black-box predictions via influence functions","author":"koh","year":"2017","journal-title":"Proc ICML"},{"key":"ref24","article-title":"Adversarial examples in the physical world","author":"kurakin","year":"2016","journal-title":"arXiv 1607 02533"},{"key":"ref23","first-page":"1097","article-title":"ImageNet classification with deep convolutional neural networks","author":"krizhevsky","year":"2012","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref26","first-page":"7167","article-title":"A simple unified framework for detecting out-of-distribution samples and adversarial attacks","author":"lee","year":"2018","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref25","first-page":"598","article-title":"Optimal brain damage","author":"lecun","year":"1990","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1002\/isaf.289"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00928"},{"key":"ref58","article-title":"Visualizing deep neural network decisions: Prediction difference analysis","author":"zintgraf","year":"2017","journal-title":"arXiv 1702 04595"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.378"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.469"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00920"},{"key":"ref54","first-page":"543","article-title":"Top-down neural attention by excitation backprop","author":"zhang","year":"2016","journal-title":"Proc Eur Conf Comput Vis"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00922"},{"key":"ref52","article-title":"Google’s neural machine translation system: Bridging the gap between human and machine translation","author":"wu","year":"2016","journal-title":"arXiv 1609 08144"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1214\/aos\/1013203451"},{"key":"ref11","article-title":"Explaining and harnessing adversarial examples","author":"goodfellow","year":"2014","journal-title":"arXiv 1412 6572"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.74"},{"key":"ref12","first-page":"1","article-title":"Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding","author":"han","year":"2016","journal-title":"Proc ICLR"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.4310\/SII.2009.v2.n3.a8"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.155"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46493-0_1"},{"key":"ref17","article-title":"Distilling the knowledge in a neural network","author":"hinton","year":"2015","journal-title":"ArXiv 1503 02531"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.93"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.320"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.338"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1023\/A:1010933404324"},{"key":"ref6","first-page":"6967","article-title":"Real time image saliency for black box classifiers","author":"dabkowski","year":"2017","journal-title":"Proc NIPS"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1145\/3128572.3140444"},{"key":"ref8","article-title":"A roadmap for a rigorous science of interpretability","author":"doshi-velez","year":"2017","journal-title":"arXiv 1702 08608"},{"key":"ref7","article-title":"Towards interpretable deep neural networks by leveraging adversarial examples","author":"dong","year":"2017","journal-title":"arXiv 1708 05493"},{"key":"ref49","article-title":"Ensemble adversarial training: Attacks and defenses","author":"tram\u00e8r","year":"2017","journal-title":"arXiv 1705 07204"},{"key":"ref9","article-title":"Detecting adversarial samples from artifacts","author":"feinman","year":"2017","journal-title":"arXiv 1703 00410"},{"key":"ref46","first-page":"3319","article-title":"Axiomatic attribution for deep networks","author":"sundararajan","year":"2017","journal-title":"Proc ICML"},{"key":"ref45","article-title":"Striving for simplicity: The all convolutional net","author":"springenberg","year":"2014","journal-title":"arXiv 1412 6806"},{"key":"ref48","article-title":"Intriguing properties of neural networks","author":"szegedy","year":"2013","journal-title":"arXiv 1312 6199"},{"key":"ref47","first-page":"3104","article-title":"Sequence to sequence learning with neural networks","author":"sutskever","year":"2014","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref42","article-title":"Deep inside convolutional networks: Visualising image classification models and saliency maps","author":"simonyan","year":"2013","journal-title":"arXiv 1312 6034"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1038\/nature24270"},{"key":"ref44","article-title":"SmoothGrad: Removing noise by adding noise","author":"smilkov","year":"2017","journal-title":"arXiv 1706 03825"},{"key":"ref43","article-title":"Very deep convolutional networks for large-scale image recognition","author":"simonyan","year":"2014","journal-title":"arXiv 1409 1556"}],"container-title":["IEEE Transactions on Image Processing"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/83\/8835130\/09093114.pdf?arnumber=9093114","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,4,27]],"date-time":"2022-04-27T14:39:01Z","timestamp":1651070341000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9093114\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020]]},"references-count":58,"URL":"https:\/\/doi.org\/10.1109\/tip.2020.2993098","relation":{},"ISSN":["1057-7149","1941-0042"],"issn-type":[{"value":"1057-7149","type":"print"},{"value":"1941-0042","type":"electronic"}],"subject":[],"published":{"date-parts":[[2020]]}}}