{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,2]],"date-time":"2025-11-02T00:59:19Z","timestamp":1762045159909,"version":"build-2065373602"},"reference-count":83,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"4","license":[{"start":{"date-parts":[[2022,8,1]],"date-time":"2022-08-01T00:00:00Z","timestamp":1659312000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2022,8,1]],"date-time":"2022-08-01T00:00:00Z","timestamp":1659312000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2022,8,1]],"date-time":"2022-08-01T00:00:00Z","timestamp":1659312000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"IBM AI Horizon Scholarship","award":["R01EB026646","R01CA233888","R01CA237267","R01HL151561","R21CA264772","R01EB031102"],"award-info":[{"award-number":["R01EB026646","R01CA233888","R01CA237267","R01HL151561","R21CA264772","R01EB031102"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Artif. Intell."],"published-print":{"date-parts":[[2022,8]]},"DOI":"10.1109\/tai.2021.3128132","type":"journal-article","created":{"date-parts":[[2021,11,15]],"date-time":"2021-11-15T17:42:09Z","timestamp":1636998129000},"page":"595-608","source":"Crossref","is-referenced-by-count":10,"title":["On a Sparse Shortcut Topology of Artificial Neural Networks"],"prefix":"10.1109","volume":"3","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-3691-5141","authenticated-orcid":false,"given":"Feng-Lei","family":"Fan","sequence":"first","affiliation":[{"name":"Department of Biomedical Engineering, Rensselaer Polytechnic Institute, Troy, NY, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0531-2110","authenticated-orcid":false,"given":"Dayang","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering, University of Massachusetts, Lowell, MA, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4734-026X","authenticated-orcid":false,"given":"Hengtao","family":"Guo","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering, Rensselaer Polytechnic Institute, Troy, NY, USA"}]},{"given":"Qikui","family":"Zhu","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering, Rensselaer Polytechnic Institute, Troy, NY, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9779-2141","authenticated-orcid":false,"given":"Pingkun","family":"Yan","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering, Rensselaer Polytechnic Institute, Troy, NY, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2656-7705","authenticated-orcid":false,"given":"Ge","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering, Rensselaer Polytechnic Institute, Troy, NY, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5852-0813","authenticated-orcid":false,"given":"Hengyong","family":"Yu","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering, University of Massachusetts, Lowell, MA, USA"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1038\/nature14539"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/TASL.2011.2134090"},{"key":"ref3","first-page":"1378","article-title":"Ask me anything: Dynamic memory networks for natural language processing","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Kumar","year":"2016"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2017.2715284"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2016.2624938"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2016.2535865"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1145\/3065386"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.308"},{"key":"ref9","first-page":"1","article-title":"Network in network","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Lin","year":"2014"},{"key":"ref10","first-page":"398","article-title":"Very deep convolutional networks for large-scale image recognition","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Simonyan","year":"2015"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.243"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/ICPR48806.2021.9413080"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298642"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2016.2644615"},{"key":"ref16","first-page":"2377","article-title":"Training very deep networks","volume-title":"Proc. 28th Int. Conf. Neural Inf. Process. Syst.","volume":"2","author":"Srivastava","year":"2015"},{"key":"ref17","first-page":"1","article-title":"FractalNet: Ultra-deep neural networks without residuals","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Larsson","year":"2017"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1016\/0893-6080(89)90003-8"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1016\/0893-6080(89)90020-8"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2013.2296046"},{"key":"ref21","first-page":"1","article-title":"The power of deeper networks for expressing natural functions","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Rolnick","year":"2018"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1142\/S0219530516400042"},{"key":"ref23","first-page":"907","article-title":"The power of depth for feedforward neural networks","volume-title":"Proc. Conf. Learn. Theory","author":"Eldan","year":"2016"},{"key":"ref24","first-page":"1","article-title":"Why deep neural networks for function approximation","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Liang","year":"2017"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1007\/978-94-011-3030-1_56"},{"key":"ref26","first-page":"6232","article-title":"The expressive power of neural networks: A view from the width","volume-title":"Proc. 31st Int. Conf. Neural Inf. Process. Syst.","author":"Lu","year":"2017"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2020.01.007"},{"key":"ref28","first-page":"6169","article-title":"ResNet with one-neuron hidden layers is a universal approximator","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","volume":"31","author":"Lin","year":"2018"},{"key":"ref29","first-page":"550","article-title":"Residual networks behave like ensembles of relatively shallow networks","volume-title":"Proc. 30th Int. Conf. Neural Inf. Process. Syst.","author":"Veit","year":"2016"},{"key":"ref30","first-page":"7892","article-title":"Towards understanding the importance of shortcut connections in residual networks","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","author":"Liu","year":"2019"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2020.2966319"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13284"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2019.2922960"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1016\/0167-8396(94)90004-3"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01258-8_12"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2021.01.020"},{"key":"ref37","first-page":"6241","article-title":"Spectrally-normalized margin bounds for neural networks","volume-title":"Proc. 31st Int. Conf. Neural Inf. Process. Syst.","author":"Bartlett","year":"2017"},{"key":"ref38","first-page":"1376","article-title":"Norm-based capacity control in neural networks","volume-title":"Proc. Conf. Learn. Theory","author":"Neyshabur","year":"2015"},{"key":"ref39","first-page":"1","article-title":"A PAC-Bayesian approach to spectrally-normalized margin bounds for neural networks","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Neyshabur","year":"2018"},{"key":"ref40","first-page":"8289","article-title":"How SGD selects the global minima in over-parameterized learning: A dynamical stability perspective","volume-title":"Proc. 32nd Int. Conf. Neural Inf. Process. Syst.","author":"Wu","year":"2018"},{"key":"ref41","first-page":"6158","article-title":"Learning and generalization in overparameterized neural networks, going beyond two layers","volume-title":"Proc. 33rd Int. Conf. Neural Information Proc. Syst.","author":"Allen-Zhu","year":"2019"},{"key":"ref42","first-page":"1","article-title":"SGD learns over-parameterized networks that provably generalize on linearly separable data","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Brutzkus","year":"2018"},{"key":"ref43","first-page":"1","article-title":"The role of over-parametrization in generalization of neural networks","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Neyshabur","year":"2018"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1088\/1742-5468\/ac3a74"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-41136-6_11"},{"key":"ref46","first-page":"8571","article-title":"Neural tangent kernel: Convergence and generalization in neural networks","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","author":"Jacot","year":"2018"},{"key":"ref47","first-page":"8141","article-title":"On exact computation with an infinitely wide neural net","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","author":"Arora","year":"2019"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1016\/0095-0696(78)90006-2"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1007\/s13398-014-0173-7.2"},{"issue":"1","key":"ref50","first-page":"1929","article-title":"Dropout: A simple way to prevent neural networks from overfitting","volume":"15","author":"Srivastava","year":"2014","journal-title":"J. Mach. Learn. Res."},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46493-0_39"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i04.5842"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/TAI.2022.3185179"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2023.3328829"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00474"},{"key":"ref56","first-page":"6105","article-title":"EfficientNet: Rethinking model scaling for convolutional neural networks","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Tan","year":"2019"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00353"},{"key":"ref58","first-page":"1","article-title":"LambdaNetworks: Modeling long-range interactions without attention","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Bello","year":"2020"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00745"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01179"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00165"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00137"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00716"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00907"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33014780"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01246-5_2"},{"key":"ref68","first-page":"1","article-title":"DARTS: Differentiable architecture search","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Liu","year":"2018"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01099"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01044"},{"key":"ref71","first-page":"10347","article-title":"Training data-efficient image transformers & distillation through attention","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Touvron","year":"2021"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00188"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01064"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01155"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.1109\/TRPMS.2021.3066428"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2020.2967051"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1016\/j.inffus.2019.12.012"},{"key":"ref78","first-page":"4124","article-title":"Full-gradient representation for neural network visualization","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","author":"Srinivas","year":"2019"},{"key":"ref79","article-title":"SmoothGrad: Removing noise by adding noise","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Smilkov","year":"2017"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.1007\/s10144-009-0162-4"},{"key":"ref81","first-page":"1","article-title":"SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and $< $0.5 MB model size","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Iandola","year":"2017"},{"key":"ref82","doi-asserted-by":"publisher","DOI":"10.2307\/1932409"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2018\/391"}],"container-title":["IEEE Transactions on Artificial Intelligence"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9078688\/9834985\/09614997.pdf?arnumber=9614997","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,23]],"date-time":"2025-08-23T01:10:07Z","timestamp":1755911407000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9614997\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,8]]},"references-count":83,"journal-issue":{"issue":"4"},"URL":"https:\/\/doi.org\/10.1109\/tai.2021.3128132","relation":{},"ISSN":["2691-4581"],"issn-type":[{"type":"electronic","value":"2691-4581"}],"subject":[],"published":{"date-parts":[[2022,8]]}}}