{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T05:18:47Z","timestamp":1788239927930,"version":"build-2803163510"},"reference-count":53,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"9","license":[{"start":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T00:00:00Z","timestamp":1788220800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T00:00:00Z","timestamp":1788220800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T00:00:00Z","timestamp":1788220800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"SERB Start-up Research Grant (SRG), Government of India"},{"name":"Kotak AI-ML Center (KIAC) Ph.D. Fellowship"},{"name":"Prime Minister\u2019s Research Fellowship (PMRF), India"},{"name":"Indian Institute of Science (IISc), Bengaluru, India","award":["Arcot Ramachandran Young Investigator Award"],"award-info":[{"award-number":["Arcot Ramachandran Young Investigator Award"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Artif. Intell."],"published-print":{"date-parts":[[2026,9]]},"DOI":"10.1109\/tai.2026.3656866","type":"journal-article","created":{"date-parts":[[2026,1,22]],"date-time":"2026-01-22T21:05:16Z","timestamp":1769115916000},"page":"4806-4821","source":"Crossref","is-referenced-by-count":0,"title":["On Convergence of Adam With Data Dependent Stepsize"],"prefix":"10.1109","volume":"7","author":[{"ORCID":"https:\/\/orcid.org\/0009-0002-7543-0649","authenticated-orcid":false,"given":"Alokendu","family":"Mazumder","sequence":"first","affiliation":[{"name":"Indian Institute of Science (IISc)","place":["Bengaluru, India"],"department":["Robert Bosch Center for Cyber-Physical Systems (RBCCPS)"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-6055-4246","authenticated-orcid":false,"given":"Rishab","family":"Sabharwal","sequence":"additional","affiliation":[{"name":"Indian Institute of Science (IISc)","place":["Bengaluru, India"],"department":["Robert Bosch Center for Cyber-Physical Systems (RBCCPS)"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-4456-0165","authenticated-orcid":false,"given":"Bhartendu","family":"Kumar","sequence":"additional","affiliation":[{"name":"Indian Institute of Science (IISc)","place":["Bengaluru, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-2225-5941","authenticated-orcid":false,"given":"Manan","family":"Tayal","sequence":"additional","affiliation":[{"name":"Indian Institute of Science (IISc)","place":["Bengaluru, India"],"department":["Robert Bosch Center for Cyber-Physical Systems (RBCCPS)"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-0201-3086","authenticated-orcid":false,"given":"Chirag","family":"Garg","sequence":"additional","affiliation":[{"name":"Indian Institute of Science (IISc)","place":["Bengaluru, India"],"department":["Robert Bosch Center for Cyber-Physical Systems (RBCCPS)"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-8814-1589","authenticated-orcid":false,"given":"Arnab","family":"Roy","sequence":"additional","affiliation":[{"name":"Indian Institute of Science (IISc)","place":["Bengaluru, India"],"department":["Robert Bosch Center for Cyber-Physical Systems (RBCCPS)"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4835-4556","authenticated-orcid":false,"given":"Punit","family":"Rathore","sequence":"additional","affiliation":[{"name":"Indian Institute of Science (IISc)","place":["Bengaluru, India"],"department":["Robert Bosch Center for Cyber-Physical Systems (RBCCPS)"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-35289-8_25"},{"key":"ref2","first-page":"2624","article-title":"Curiously fast convergence of some stochastic gradient descent algorithms","volume-title":"Proc. Symp. Learn. Data Sci.","author":"Bottou","year":"2009"},{"issue":"7","key":"ref3","article-title":"Adaptive subgradient methods for online learning and stochastic optimization","volume":"12","author":"Duchi","year":"2011","journal-title":"J. Mach. Learn. Res."},{"key":"ref4","article-title":"RMSProp: Divide the gradient by a running average of its recent magnitude. Coursera: Neural networks for machine learning","volume-title":"Proc. COURSERA Neural Netw. Mach. Learn","volume":"17","author":"Tieleman","year":"2012"},{"key":"ref5","first-page":"1","article-title":"Adam: A method for stochastic optimization","volume-title":"Proc. 3rd Int. Conf. Learn. Represent.","author":"Kingma","year":"2015"},{"key":"ref6","article-title":"Adadelta: An adaptive learning rate method","author":"Zeiler","year":"2012"},{"key":"ref7","first-page":"1195","article-title":"Fast and faster convergence of SGD for over-parameterized models and an accelerated perceptron","volume-title":"Proc. 22nd AISTATS","author":"Vaswani","year":"2019"},{"key":"ref8","article-title":"On the convergence of Adam and beyond","volume-title":"Proc. ICLR","author":"Sashank","year":"2018"},{"key":"ref9","article-title":"Optimizing neural networks with Kronecker-factored approximate curvature","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Martens","year":"2015"},{"key":"ref10","article-title":"Automated inference with adaptive batches","volume-title":"Proc. Artif. Intell. Statist.","author":"De","year":"2017"},{"key":"ref11","first-page":"2251","article-title":"Stopwasting my gradients: Practical SVRG","volume":"28","author":"Babanezhad Harikandeh","year":"2015","journal-title":"Advances in Neural Information Processing System"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.5555\/2999611.2999647"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.5555\/2968826.2969010"},{"key":"ref14","first-page":"560","article-title":"signSGD: Compressed optimisation for non-convex problems","volume-title":"Proc. Int. Conf. Mach. Learn","author":"Bernstein","year":"2018"},{"key":"ref15","first-page":"225","article-title":"Convergence rates of a momentum algorithm with bounded adaptive step size for nonconvex optimization","volume-title":"Proc. 12th Asian Conf. Mach. Learn., Ser. Proc. Mach. Learn. Res.","volume":"129","author":"Barakat","year":"2020"},{"key":"ref16","article-title":"RMSProp converges with proper hyper-parameter","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Shi","year":"2020"},{"key":"ref17","article-title":"Amos: An Adam-style optimizer with adaptive weight decay towards model-oriented scale","author":"Tian","year":"2022"},{"key":"ref18","article-title":"On the convergence of a class of Adam-type algorithms for non-convex optimization","author":"Chen","year":"2018"},{"key":"ref19","article-title":"Adaptive gradient methods with dynamic bound of learning rate","author":"Luo","year":"2019"},{"key":"ref20","article-title":"A simple convergence proof of Adam and Adagrad","author":"D\u00e9fossez","year":"2020"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01138"},{"key":"ref22","article-title":"AdaShift: Decorrelation and convergence of adaptive learning rate methods","author":"Zhou","year":"2018"},{"issue":"1","key":"ref23","first-page":"10411","article-title":"Towards practical Adam: Non-convexity, convergence theory, and mini-batch acceleration","volume":"23","author":"Chen","year":"2022","journal-title":"J. Mach. Learn. Res."},{"key":"ref24","first-page":"14736","article-title":"Implicit rank-minimizing autoencoder","volume":"33","author":"Jing","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i10.17037"},{"key":"ref26","article-title":"Convergence guarantees for RMSProp and ADAM in non-convex optimization and an empirical comparison to Nesterov acceleration","author":"De","year":"2018"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1145\/2623330.2623612"},{"key":"ref28","article-title":"Lipschitz regularity of deep neural networks: Analysis and efficient estimation","volume-title":"Advances in Neural Information Processing Systems","author":"Virmaux","year":"2018"},{"key":"ref29","article-title":"Efficient and accurate estimation of Lipschitz constants for deep neural networks","volume-title":"Advances in Neural Information Processing Systems","author":"Fazlyab","year":"2019"},{"key":"ref30","article-title":"Lipschitz constant estimation of neural networks via sparse polynomial optimization","author":"Latorre","year":"2020"},{"key":"ref31","article-title":"Efficient and accurate estimation of Lipschitz constants for deep neural networks","volume":"32","author":"Fazlyab","year":"2019","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1007\/s10994-020-05929-w"},{"key":"ref33","volume-title":"Geometric Measure Theory","author":"Federer","year":"2014"},{"key":"ref34","first-page":"43","article-title":"Solution directe de l\u2019\u00e9quation s\u00e9culaire et de quelques problemes analogues transcendants","volume":"156","author":"M\u00fcntz","year":"1913","journal-title":"CR Acad. Sci. Paris"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511804441"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1137\/1011036"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.123"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/5.726791"},{"key":"ref39","article-title":"Very deep convolutional networks for large-scale image recognition","author":"Simonyan","year":"2014"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00474"},{"key":"ref42","article-title":"An image is worth 16x16 words: Transformers for image recognition at scale","author":"Dosovitskiy","year":"2020"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref44","first-page":"9815","article-title":"Adaptive methods for nonconvex optimization","volume":"31","author":"Zaheer","year":"2018","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref45","first-page":"404","article-title":"Dissecting Adam: The sign, magnitude and variance of stochastic gradients","volume-title":"Proc. Int. Conf. Mach. Learn","author":"Balles","year":"2018"},{"key":"ref46","article-title":"On the variance of the adaptive learning rate and beyond","author":"Liu","year":"2019"},{"key":"ref47","article-title":"Decoupled weight decay regularization","author":"Loshchilov","year":"2017"},{"key":"ref48","article-title":"The Llama 3 herd of models","author":"Dubey","year":"2024"},{"key":"ref49","article-title":"Qwen3 technical report","author":"Yang","year":"2025"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D16-1264"},{"key":"ref51","article-title":"Benchmarking neural network training algorithms","author":"Dahl","year":"2023"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2020\/452"},{"issue":"2","key":"ref53","first-page":"3","article-title":"LoRA: Low-rank adaptation of large language models","volume":"1","author":"Hu","year":"2022","journal-title":"ICLR"}],"container-title":["IEEE Transactions on Artificial Intelligence"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/9078688\/11673324\/11361385.pdf?arnumber=11361385","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T04:58:21Z","timestamp":1788238701000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11361385\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9]]},"references-count":53,"journal-issue":{"issue":"9"},"URL":"https:\/\/doi.org\/10.1109\/tai.2026.3656866","relation":{"has-preprint":[{"id-type":"doi","id":"10.36227\/techrxiv.176238048.87572362\/v1","asserted-by":"object"}]},"ISSN":["2691-4581"],"issn-type":[{"value":"2691-4581","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9]]}}}