{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,19]],"date-time":"2026-02-19T15:40:46Z","timestamp":1771515646734,"version":"3.50.1"},"reference-count":37,"publisher":"IEEE","license":[{"start":{"date-parts":[[2019,12,1]],"date-time":"2019-12-01T00:00:00Z","timestamp":1575158400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2019,12,1]],"date-time":"2019-12-01T00:00:00Z","timestamp":1575158400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2019,12,1]],"date-time":"2019-12-01T00:00:00Z","timestamp":1575158400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2019,12]]},"DOI":"10.1109\/bigdata47090.2019.9006550","type":"proceedings-article","created":{"date-parts":[[2020,2,25]],"date-time":"2020-02-25T06:05:34Z","timestamp":1582610734000},"page":"830-839","source":"Crossref","is-referenced-by-count":26,"title":["Improving Scalability of Parallel CNN Training by Adjusting Mini-Batch Size at Run-Time"],"prefix":"10.1109","author":[{"given":"Sunwoo","family":"Lee","sequence":"first","affiliation":[]},{"given":"Qiao","family":"Kang","sequence":"additional","affiliation":[]},{"given":"Sandeep","family":"Madireddy","sequence":"additional","affiliation":[]},{"given":"Prasanna","family":"Balaprakash","sequence":"additional","affiliation":[]},{"given":"Ankit","family":"Agrawal","sequence":"additional","affiliation":[]},{"given":"Alok","family":"Choudhary","sequence":"additional","affiliation":[]},{"given":"Richard","family":"Archibald","sequence":"additional","affiliation":[]},{"given":"Wei-keng","family":"Liao","sequence":"additional","affiliation":[]}],"member":"263","reference":[{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1145\/1362622.1362692"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1007\/s00607-013-0327-z"},{"key":"ref31","first-page":"1731","article-title":"Train longer, generalize better: closing the generalization gap in large batch training of neural networks","author":"hoffer","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"ref30","article-title":"Adam: A method for stochastic optimization","author":"kingma","year":"2014","journal-title":"arXiv preprint arXiv 1412 6980"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1145\/3126908.3126941"},{"key":"ref36","first-page":"659","article-title":"High-performance matrix-matrix multiplications of very small matrices","author":"masliah","year":"2016","journal-title":"European Conference on Parallel Processing"},{"key":"ref35","article-title":"Adabatch: adaptive batch sizes for training deep neural networks","author":"devarakonda","year":"2017","journal-title":"arXiv preprint arXiv 1712 02029"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1177\/1094342005054257"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.298"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW.2017.151"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1214\/aoms\/1177729586"},{"key":"ref13","article-title":"Highly scalable deep learning training system with mixed-precision: Training imagenet in four minutes","author":"jia","year":"2018","journal-title":"arXiv preprint arXiv 1807 11205"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/SC.2018.00068"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1145\/3126908.3126916"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/SC.2018.00054"},{"key":"ref17","article-title":"Don't decay the learning rate, increase the batch size","author":"smith","year":"2017","journal-title":"arXiv preprint arXiv 1711 00540"},{"key":"ref18","article-title":"On large-batch training for deep learning: Generalization gap and sharp minima","author":"keskar","year":"2016","journal-title":"arXiv preprint arXiv 1609 04802"},{"key":"ref19","first-page":"9322","article-title":"The effect of network width on the performance of large-batch training","author":"chen","year":"2018","journal-title":"Advances in neural information processing systems"},{"key":"ref28","first-page":"1097","article-title":"Imagenet classification with deep convolutional neural networks","author":"krizhevsky","year":"2012","journal-title":"Advances in neural information processing systems"},{"key":"ref4","article-title":"Compression artifacts removal using convolutional neural networks","author":"svoboda","year":"2016","journal-title":"arXiv preprint arXiv 1605 09090"},{"key":"ref27","article-title":"Deep learning","author":"goodfellow","year":"2016"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.73"},{"key":"ref6","article-title":"Image restoration using convolutional auto-encoders with symmetric skip connections","author":"mao","year":"2016","journal-title":"arXiv preprint arXiv 1606 08921"},{"key":"ref29","article-title":"Coupling adaptive batch sizes with learning rates","author":"balles","year":"2016","journal-title":"arXiv preprint arXiv 1612 05086"},{"key":"ref5","article-title":"Denoising prior driven deep neural network for image restoration","author":"dong","year":"2018","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00442"},{"key":"ref7","first-page":"2553","article-title":"Deep neural networks for object detection","author":"szegedy","year":"2013","journal-title":"Advances in neural information processing systems"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.243"},{"key":"ref9","first-page":"1646","article-title":"Accurate image super-resolution using very deep convolutional networks","author":"kim","year":"2016","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW.2017.149"},{"key":"ref22","article-title":"Scaling sgd batch size to 32k for imagenet training","author":"you","year":"2017","journal-title":"arXiv preprint arXiv 1708 02562"},{"key":"ref21","article-title":"Accurate, large minibatch sgd: training imagenet in 1 hour","author":"goyal","year":"2017","journal-title":"arXiv preprint arXiv 1706 02677"},{"key":"ref24","first-page":"1731","article-title":"Train longer, generalize better: closing the generalization gap in large batch training of neural networks","author":"hoffer","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"ref23","article-title":"Three factors influencing minima in sgd","author":"jastrzebski","year":"2017","journal-title":"arXiv preprint arXiv 1711 07128"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/5.726791"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/TNS.1974.6499235"}],"event":{"name":"2019 IEEE International Conference on Big Data (Big Data)","location":"Los Angeles, CA, USA","start":{"date-parts":[[2019,12,9]]},"end":{"date-parts":[[2019,12,12]]}},"container-title":["2019 IEEE International Conference on Big Data (Big Data)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/8986695\/9005444\/09006550.pdf?arnumber=9006550","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,7,17]],"date-time":"2022-07-17T21:46:55Z","timestamp":1658094415000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9006550\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2019,12]]},"references-count":37,"URL":"https:\/\/doi.org\/10.1109\/bigdata47090.2019.9006550","relation":{},"subject":[],"published":{"date-parts":[[2019,12]]}}}