{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,9,17]],"date-time":"2025-09-17T14:53:41Z","timestamp":1758120821657,"version":"3.28.0"},"reference-count":37,"publisher":"IEEE","license":[{"start":{"date-parts":[[2021,11,1]],"date-time":"2021-11-01T00:00:00Z","timestamp":1635724800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2021,11,1]],"date-time":"2021-11-01T00:00:00Z","timestamp":1635724800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2021,11,1]],"date-time":"2021-11-01T00:00:00Z","timestamp":1635724800000},"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":[[2021,11,1]]},"DOI":"10.1109\/iccad51958.2021.9643487","type":"proceedings-article","created":{"date-parts":[[2021,12,23]],"date-time":"2021-12-23T23:06:46Z","timestamp":1640300806000},"page":"1-9","source":"Crossref","is-referenced-by-count":5,"title":["AutoGTCO: Graph and Tensor Co-Optimize for Image Recognition with Transformers on GPU"],"prefix":"10.1109","author":[{"given":"Yang","family":"Bai","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xufeng","family":"Yao","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Qi","family":"Sun","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Bei","family":"Yu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1145\/2939672.2939785"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298965"},{"key":"ref31","first-page":"91","article-title":"Faster R-CNN: Towards real-time object detection with region proposal networks","author":"ren","year":"2015","journal-title":"Annual Conference on Neural Information Processing Systems (NIPS)"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref37","first-page":"3393","article-title":"Learning to Optimize Tensor Programs","author":"chen","year":"2018","journal-title":"Annual Conference on Neural Information Processing Systems (NIPS)"},{"key":"ref36","article-title":"Layer normalization","volume":"abs 1607 6450","author":"ba","year":"2016","journal-title":"CoRR"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1007\/11688839_16"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2020.3030548"},{"journal-title":"An image is worth 16x16 words Transformers for image recognition at scale","year":"2020","author":"dosovitskiy","key":"ref10"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/DAC18072.2020.9218700"},{"key":"ref12","article-title":"Deep neural network hardware deployment optimization via advanced active learning","author":"sun","year":"2021","journal-title":"IEEE\/ACM Proceedings Design Automation and Test in Eurpoe (DATE)"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1145\/3373376.3378508"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1145\/3316781.3317829"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2021.3084900"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/JETCAS.2019.2910232"},{"key":"ref17","first-page":"265","article-title":"TensorFlow: A system for large-scale machine learning","author":"abadi","year":"2016","journal-title":"USENIX Symposium on Operating Systems Design and Implementation (OSDI)"},{"key":"ref18","article-title":"Automatic differentiation in PyTorch","author":"paszke","year":"2017","journal-title":"NIPS Workshop"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/2647868.2654889"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1145\/3341301.3359630"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/ICME.2019.00188"},{"key":"ref27","first-page":"578","article-title":"TVM: An automated end-to-end optimizing compiler for deep learning","author":"chen","year":"2018","journal-title":"USENIX Symposium on Operating Systems Design and Implementation (OSDI)"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.308"},{"journal-title":"Transformers in vision A survey","year":"2021","author":"khan","key":"ref6"},{"key":"ref29","article-title":"IOS: Inter-operator scheduler for cnn acceleration","volume":"3","author":"ding","year":"2021","journal-title":"Machine Learning and Systems (MLSys)"},{"key":"ref5","first-page":"5998","article-title":"Attention is all you need","author":"vaswani","year":"2017","journal-title":"Annual Conference on Neural Information Processing Systems (NIPS)"},{"journal-title":"Emerging Properties in Self-Supervised Vision Transformers","year":"2021","author":"caron","key":"ref8"},{"key":"ref7","first-page":"213","article-title":"End-to-end object detection with transformers","author":"carion","year":"2020","journal-title":"European Conference on Computer Vision (ECCV)"},{"key":"ref2","first-page":"1","article-title":"Very deep convolutional networks for large-scale image recognition","author":"simonyan","year":"2015","journal-title":"International Conference on Learning Representations (ICLR)"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00681"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"journal-title":"cuDNN Efficient Primitives for Deep Learning","year":"2014","author":"chetlur","key":"ref20"},{"key":"ref22","first-page":"863","article-title":"Ansor: Generating high-performance tensor programs for deep learning","author":"zheng","year":"2020","journal-title":"USENIX Symposium on Operating Systems Design and Implementation (OSDI)"},{"journal-title":"oneAPI Deep Neural Network Library","year":"0","key":"ref21"},{"journal-title":"Nvidia tensorrt","year":"0","key":"ref24"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1145\/2185520.2185528"},{"key":"ref26","article-title":"Deep Learning using Rectified Linear Units (ReLU)","volume":"abs 1803 8375","author":"agarap","year":"2018","journal-title":"CoRR"},{"key":"ref25","first-page":"448","article-title":"Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift","author":"ioffe","year":"2015","journal-title":"International Conference on Machine Learning (ICML)"}],"event":{"name":"2021 IEEE\/ACM International Conference On Computer Aided Design (ICCAD)","start":{"date-parts":[[2021,11,1]]},"location":"Munich, Germany","end":{"date-parts":[[2021,11,4]]}},"container-title":["2021 IEEE\/ACM International Conference On Computer Aided Design (ICCAD)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9643423\/9643432\/09643487.pdf?arnumber=9643487","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,5,10]],"date-time":"2022-05-10T16:55:11Z","timestamp":1652201711000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9643487\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,11,1]]},"references-count":37,"URL":"https:\/\/doi.org\/10.1109\/iccad51958.2021.9643487","relation":{},"subject":[],"published":{"date-parts":[[2021,11,1]]}}}