{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T15:53:07Z","timestamp":1783439587610,"version":"3.54.6"},"publisher-location":"New York, NY, USA","reference-count":78,"publisher":"ACM","license":[{"start":{"date-parts":[[2022,6,27]],"date-time":"2022-06-27T00:00:00Z","timestamp":1656288000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"DOI":"10.13039\/501100012166","name":"National Key Research and Development Program of China","doi-asserted-by":"publisher","award":["2020YFB2104100"],"award-info":[{"award-number":["2020YFB2104100"]}],"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":["61725201,62172008,62102045"],"award-info":[{"award-number":["61725201,62172008,62102045"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Beijing Outstanding Young Scientist Program","award":["BJJWZYJH01201910001004"],"award-info":[{"award-number":["BJJWZYJH01201910001004"]}]},{"name":"PKU-Baidu Fund Project","award":["2020BD007"],"award-info":[{"award-number":["2020BD007"]}]},{"DOI":"10.13039\/501100005090","name":"Beijing Nova Program","doi-asserted-by":"publisher","award":["Z211100002121118"],"award-info":[{"award-number":["Z211100002121118"]}],"id":[{"id":"10.13039\/501100005090","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2022,6,27]]},"DOI":"10.1145\/3498361.3538928","type":"proceedings-article","created":{"date-parts":[[2022,6,16]],"date-time":"2022-06-16T16:21:53Z","timestamp":1655396513000},"page":"450-463","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":53,"title":["Melon"],"prefix":"10.1145","author":[{"given":"Qipeng","family":"Wang","sequence":"first","affiliation":[{"name":"Peking University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mengwei","family":"Xu","sequence":"additional","affiliation":[{"name":"BUPT, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chao","family":"Jin","sequence":"additional","affiliation":[{"name":"Peking University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xinran","family":"Dong","sequence":"additional","affiliation":[{"name":"Peking University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinliang","family":"Yuan","sequence":"additional","affiliation":[{"name":"BUPT, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xin","family":"Jin","sequence":"additional","affiliation":[{"name":"Peking University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gang","family":"Huang","sequence":"additional","affiliation":[{"name":"Peking University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yunxin","family":"Liu","sequence":"additional","affiliation":[{"name":"Tsinghua University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xuanzhe","family":"Liu","sequence":"additional","affiliation":[{"name":"Peking University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2022,6,27]]},"reference":[{"key":"e_1_3_2_1_1_1","volume-title":"Collaborative machine learning without centralized training data. https:\/\/ai.googleblog.com\/2017\/04\/federated-learning-collaborative.html","author":"Federated","year":"2017","unstructured":"Federated learning : Collaborative machine learning without centralized training data. https:\/\/ai.googleblog.com\/2017\/04\/federated-learning-collaborative.html , 2017 . Federated learning: Collaborative machine learning without centralized training data. https:\/\/ai.googleblog.com\/2017\/04\/federated-learning-collaborative.html, 2017."},{"key":"e_1_3_2_1_2_1","volume-title":"https:\/\/www.technologyreview.com\/2019\/12\/11\/131629\/apple-ai-personalizes-siri-federated-learning\/","author":"How","year":"2019","unstructured":"How apple personalizes siri without hoovering up your data. https:\/\/www.technologyreview.com\/2019\/12\/11\/131629\/apple-ai-personalizes-siri-federated-learning\/ , 2019 . How apple personalizes siri without hoovering up your data. https:\/\/www.technologyreview.com\/2019\/12\/11\/131629\/apple-ai-personalizes-siri-federated-learning\/, 2019."},{"key":"e_1_3_2_1_3_1","volume-title":"https:\/\/deeplearning4j.org\/","author":"Deep","year":"2021","unstructured":"Deep learning for java. https:\/\/deeplearning4j.org\/ , 2021 . Deep learning for java. https:\/\/deeplearning4j.org\/, 2021."},{"key":"e_1_3_2_1_4_1","volume-title":"https:\/\/gdpr-info.eu\/","author":"General","year":"2021","unstructured":"General data protection regulation. https:\/\/gdpr-info.eu\/ , 2021 . Accessed Dec 5, 2021. General data protection regulation. https:\/\/gdpr-info.eu\/, 2021. Accessed Dec 5, 2021."},{"key":"e_1_3_2_1_5_1","volume-title":"https:\/\/www.tensorflow.org\/lite\/examples\/on_device_training\/overview","year":"2021","unstructured":"On-device training with tensorflow lite. https:\/\/www.tensorflow.org\/lite\/examples\/on_device_training\/overview , 2021 . On-device training with tensorflow lite. https:\/\/www.tensorflow.org\/lite\/examples\/on_device_training\/overview, 2021."},{"key":"e_1_3_2_1_6_1","first-page":"265","volume-title":"et al. Tensorflow: A system for large-scale machine learning. In <u>OSDI<\/u>","author":"Abadi Mart\u00edn","year":"2016","unstructured":"Mart\u00edn Abadi , Paul Barham , Jianmin Chen , Zhifeng Chen , Andy Davis , Jeffrey Dean , Matthieu Devin , Sanjay Ghemawat , Geoffrey Irving , Michael Isard , et al. Tensorflow: A system for large-scale machine learning. In <u>OSDI<\/u> , pages 265 -- 283 , 2016 . Mart\u00edn Abadi, Paul Barham, Jianmin Chen, Zhifeng Chen, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Geoffrey Irving, Michael Isard, et al. Tensorflow: A system for large-scale machine learning. In <u>OSDI<\/u>, pages 265--283, 2016."},{"key":"e_1_3_2_1_7_1","first-page":"1","volume-title":"Fused-layer cnn accelerators. In <u>2016 49th Annual IEEE\/ACM International Symposium on Microarchitecture (MICRO)<\/u>","author":"Alwani Manoj","year":"2016","unstructured":"Manoj Alwani , Han Chen , Michael Ferdman , and Peter Milder . Fused-layer cnn accelerators. In <u>2016 49th Annual IEEE\/ACM International Symposium on Microarchitecture (MICRO)<\/u> , pages 1 -- 12 . IEEE , 2016 . Manoj Alwani, Han Chen, Michael Ferdman, and Peter Milder. Fused-layer cnn accelerators. In <u>2016 49th Annual IEEE\/ACM International Symposium on Microarchitecture (MICRO)<\/u>, pages 1--12. IEEE, 2016."},{"key":"e_1_3_2_1_8_1","volume-title":"A genetic algorithm for the two-dimensional strip packing problem with rectangular pieces. <u>European Journal of Operational Research<\/u>, 172(3):814--837","author":"Bortfeldt Andreas","year":"2006","unstructured":"Andreas Bortfeldt . A genetic algorithm for the two-dimensional strip packing problem with rectangular pieces. <u>European Journal of Operational Research<\/u>, 172(3):814--837 , 2006 . Andreas Bortfeldt. A genetic algorithm for the two-dimensional strip packing problem with rectangular pieces. <u>European Journal of Operational Research<\/u>, 172(3):814--837, 2006."},{"key":"e_1_3_2_1_9_1","first-page":"177","volume-title":"Large-scale machine learning with stochastic gradient descent. In <u>Proceedings of COMPSTAT'2010<\/u>","author":"Bottou L\u00e9on","year":"2010","unstructured":"L\u00e9on Bottou . Large-scale machine learning with stochastic gradient descent. In <u>Proceedings of COMPSTAT'2010<\/u> , pages 177 -- 186 . Springer , 2010 . L\u00e9on Bottou. Large-scale machine learning with stochastic gradient descent. In <u>Proceedings of COMPSTAT'2010<\/u>, pages 177--186. Springer, 2010."},{"key":"e_1_3_2_1_10_1","first-page":"31","volume-title":"Towards ubiquitous learning: A first measurement of on-device training performance. In <u>Proceedings of the 5th International Workshop on Embedded and Mobile Deep Learning<\/u>","author":"Cai Dongqi","year":"2021","unstructured":"Dongqi Cai , Qipeng Wang , Yuanqiang Liu , Yunxin Liu , Shangguang Wang , and Mengwei Xu . Towards ubiquitous learning: A first measurement of on-device training performance. In <u>Proceedings of the 5th International Workshop on Embedded and Mobile Deep Learning<\/u> , pages 31 -- 36 , 2021 . Dongqi Cai, Qipeng Wang, Yuanqiang Liu, Yunxin Liu, Shangguang Wang, and Mengwei Xu. Towards ubiquitous learning: A first measurement of on-device training performance. In <u>Proceedings of the 5th International Workshop on Embedded and Mobile Deep Learning<\/u>, pages 31--36, 2021."},{"key":"e_1_3_2_1_11_1","volume-title":"Training deep nets with sublinear memory cost. <u>arXiv preprint arXiv:1604.06174<\/u>","author":"Chen Tianqi","year":"2016","unstructured":"Tianqi Chen , Bing Xu , Chiyuan Zhang , and Carlos Guestrin . Training deep nets with sublinear memory cost. <u>arXiv preprint arXiv:1604.06174<\/u> , 2016 . Tianqi Chen, Bing Xu, Chiyuan Zhang, and Carlos Guestrin. Training deep nets with sublinear memory cost. <u>arXiv preprint arXiv:1604.06174<\/u>, 2016."},{"key":"e_1_3_2_1_12_1","first-page":"3123","volume-title":"Binaryconnect: Training deep neural networks with binary weights during propagations. In <u>Advances in neural information processing systems<\/u>","author":"Courbariaux Matthieu","year":"2015","unstructured":"Matthieu Courbariaux , Yoshua Bengio , and Jean-Pierre David . Binaryconnect: Training deep neural networks with binary weights during propagations. In <u>Advances in neural information processing systems<\/u> , pages 3123 -- 3131 , 2015 . Matthieu Courbariaux, Yoshua Bengio, and Jean-Pierre David. Binaryconnect: Training deep neural networks with binary weights during propagations. In <u>Advances in neural information processing systems<\/u>, pages 3123--3131, 2015."},{"key":"e_1_3_2_1_13_1","volume-title":"Binarized neural networks: Training deep neural networks with weights and activations constrained to+ 1 or-1. <u>arXiv preprint arXiv:1602.02830<\/u>","author":"Courbariaux Matthieu","year":"2016","unstructured":"Matthieu Courbariaux , Itay Hubara , Daniel Soudry , Ran El-Yaniv , and Yoshua Bengio . Binarized neural networks: Training deep neural networks with weights and activations constrained to+ 1 or-1. <u>arXiv preprint arXiv:1602.02830<\/u> , 2016 . Matthieu Courbariaux, Itay Hubara, Daniel Soudry, Ran El-Yaniv, and Yoshua Bengio. Binarized neural networks: Training deep neural networks with weights and activations constrained to+ 1 or-1. <u>arXiv preprint arXiv:1602.02830<\/u>, 2016."},{"key":"e_1_3_2_1_14_1","doi-asserted-by":"crossref","unstructured":"Long Duong Trevor Cohn Steven Bird and Paul Cook. Low resource dependency parsing: Cross-lingual parameter sharing in a neural network parser. In <u>Proceedings of the 53rd annual meeting of the Association for Computational Linguistics and the 7th international joint conference on natural language processing (volume 2: short papers)<\/u> pages 845--850 2015. Long Duong Trevor Cohn Steven Bird and Paul Cook. Low resource dependency parsing: Cross-lingual parameter sharing in a neural network parser. In <u>Proceedings of the 53rd annual meeting of the Association for Computational Linguistics and the 7th international joint conference on natural language processing (volume 2: short papers)<\/u> pages 845--850 2015.","DOI":"10.3115\/v1\/P15-2139"},{"key":"e_1_3_2_1_15_1","doi-asserted-by":"crossref","unstructured":"Biyi Fang Xiao Zeng and Mi Zhang. Nestdnn: Resource-aware multi-tenant on-device deep learning for continuous mobile vision. In <u>Proceedings of the 24th Annual International Conference on Mobile Computing and Networking<\/u> pages 115--127 2018. Biyi Fang Xiao Zeng and Mi Zhang. Nestdnn: Resource-aware multi-tenant on-device deep learning for continuous mobile vision. In <u>Proceedings of the 24th Annual International Conference on Mobile Computing and Networking<\/u> pages 115--127 2018.","DOI":"10.1145\/3241539.3241559"},{"key":"e_1_3_2_1_16_1","volume-title":"large minibatch sgd: Training imagenet in 1 hour. <u>arXiv preprint arXiv:1706.02677<\/u>","author":"Goyal Priya","year":"2017","unstructured":"Priya Goyal , Piotr Doll\u00e1r , Ross Girshick , Pieter Noordhuis , Lukasz Wesolowski , Aapo Kyrola , Andrew Tulloch , Yangqing Jia , and Kaiming He. Accurate , large minibatch sgd: Training imagenet in 1 hour. <u>arXiv preprint arXiv:1706.02677<\/u> , 2017 . Priya Goyal, Piotr Doll\u00e1r, Ross Girshick, Pieter Noordhuis, Lukasz Wesolowski, Aapo Kyrola, Andrew Tulloch, Yangqing Jia, and Kaiming He. Accurate, large minibatch sgd: Training imagenet in 1 hour. <u>arXiv preprint arXiv:1706.02677<\/u>, 2017."},{"key":"e_1_3_2_1_17_1","volume-title":"Memory-efficient backpropagation through time. <u>Advances in Neural Information Processing Systems<\/u>, 29:4125--4133","author":"Gruslys Audrunas","year":"2016","unstructured":"Audrunas Gruslys , R\u00e9mi Munos , Ivo Danihelka , Marc Lanctot , and Alex Graves . Memory-efficient backpropagation through time. <u>Advances in Neural Information Processing Systems<\/u>, 29:4125--4133 , 2016 . Audrunas Gruslys, R\u00e9mi Munos, Ivo Danihelka, Marc Lanctot, and Alex Graves. Memory-efficient backpropagation through time. <u>Advances in Neural Information Processing Systems<\/u>, 29:4125--4133, 2016."},{"key":"e_1_3_2_1_18_1","volume-title":"PMLR","author":"Gupta Suyog","year":"2015","unstructured":"Suyog Gupta , Ankur Agrawal , Kailash Gopalakrishnan , and Pritish Narayanan . Deep learning with limited numerical precision. In <u>International conference on machine learning<\/u>, pages 1737--1746 . PMLR , 2015 . Suyog Gupta, Ankur Agrawal, Kailash Gopalakrishnan, and Pritish Narayanan. Deep learning with limited numerical precision. In <u>International conference on machine learning<\/u>, pages 1737--1746. PMLR, 2015."},{"key":"e_1_3_2_1_19_1","first-page":"770","volume-title":"Deep residual learning for image recognition. In <u>Proceedings of the IEEE conference on computer vision and pattern recognition<\/u>","author":"He Kaiming","year":"2016","unstructured":"Kaiming He , Xiangyu Zhang , Shaoqing Ren , and Jian Sun . Deep residual learning for image recognition. In <u>Proceedings of the IEEE conference on computer vision and pattern recognition<\/u> , pages 770 -- 778 , 2016 . Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In <u>Proceedings of the IEEE conference on computer vision and pattern recognition<\/u>, pages 770--778, 2016."},{"key":"e_1_3_2_1_20_1","first-page":"2","volume-title":"Montr\u00e9al, Canada<\/u>","author":"Xiaoxi HE","year":"2018","unstructured":"Xiaoxi HE , Zimu ZHOU, and Lothar THIELE. Multi-task zipping via layer-wise neuron sharing.(2018). In <u>Proceedings of the 32nd Annual Conference on Advances in Neural Information Processing Systems , Montr\u00e9al, Canada<\/u> , pages 2 -- 8 , 2018 . Xiaoxi HE, Zimu ZHOU, and Lothar THIELE. Multi-task zipping via layer-wise neuron sharing.(2018). In <u>Proceedings of the 32nd Annual Conference on Advances in Neural Information Processing Systems, Montr\u00e9al, Canada<\/u>, pages 2--8, 2018."},{"key":"e_1_3_2_1_21_1","volume-title":"Train longer, generalize better: closing the generalization gap in large batch training of neural networks. <u>arXiv preprint arXiv:1705.08741<\/u>","author":"Hoffer Elad","year":"2017","unstructured":"Elad Hoffer , Itay Hubara , and Daniel Soudry . Train longer, generalize better: closing the generalization gap in large batch training of neural networks. <u>arXiv preprint arXiv:1705.08741<\/u> , 2017 . Elad Hoffer, Itay Hubara, and Daniel Soudry. Train longer, generalize better: closing the generalization gap in large batch training of neural networks. <u>arXiv preprint arXiv:1705.08741<\/u>, 2017."},{"key":"e_1_3_2_1_22_1","volume-title":"Mobilenets: Efficient convolutional neural networks for mobile vision applications. <u>arXiv preprint arXiv:1704.04861<\/u>","author":"Howard Andrew G","year":"2017","unstructured":"Andrew G Howard , Menglong Zhu , Bo Chen , Dmitry Kalenichenko , Weijun Wang , Tobias Weyand , Marco Andreetto , and Hartwig Adam . Mobilenets: Efficient convolutional neural networks for mobile vision applications. <u>arXiv preprint arXiv:1704.04861<\/u> , 2017 . Andrew G Howard, Menglong Zhu, Bo Chen, Dmitry Kalenichenko, Weijun Wang, Tobias Weyand, Marco Andreetto, and Hartwig Adam. Mobilenets: Efficient convolutional neural networks for mobile vision applications. <u>arXiv preprint arXiv:1704.04861<\/u>, 2017."},{"key":"e_1_3_2_1_23_1","first-page":"1341","volume-title":"Swapadvisor: Pushing deep learning beyond the gpu memory limit via smart swapping. In <u>ASPLOS<\/u>","author":"Huang Chien-Chin","year":"2020","unstructured":"Chien-Chin Huang , Gu Jin , and Jinyang Li . Swapadvisor: Pushing deep learning beyond the gpu memory limit via smart swapping. In <u>ASPLOS<\/u> , pages 1341 -- 1355 , 2020 . Chien-Chin Huang, Gu Jin, and Jinyang Li. Swapadvisor: Pushing deep learning beyond the gpu memory limit via smart swapping. In <u>ASPLOS<\/u>, pages 1341--1355, 2020."},{"key":"e_1_3_2_1_24_1","volume-title":"et al. Gpipe: Efficient training of giant neural networks using pipeline parallelism. <u>Advances in neural information processing systems<\/u>, 32:103--112","author":"Huang Yanping","year":"2019","unstructured":"Yanping Huang , Youlong Cheng , Ankur Bapna , Orhan Firat , Dehao Chen , Mia Chen , HyoukJoong Lee , Jiquan Ngiam , Quoc V Le , Yonghui Wu , et al. Gpipe: Efficient training of giant neural networks using pipeline parallelism. <u>Advances in neural information processing systems<\/u>, 32:103--112 , 2019 . Yanping Huang, Youlong Cheng, Ankur Bapna, Orhan Firat, Dehao Chen, Mia Chen, HyoukJoong Lee, Jiquan Ngiam, Quoc V Le, Yonghui Wu, et al. Gpipe: Efficient training of giant neural networks using pipeline parallelism. <u>Advances in neural information processing systems<\/u>, 32:103--112, 2019."},{"key":"e_1_3_2_1_25_1","volume-title":"Squeezenet: Alexnet-level accuracy with 50x fewer parameters and &lt","author":"Iandola Forrest N","year":"2016","unstructured":"Forrest N Iandola , Song Han , Matthew W Moskewicz , Khalid Ashraf , William J Dally , and Kurt Keutzer . Squeezenet: Alexnet-level accuracy with 50x fewer parameters and &lt ; 0.5 mb model size. <u>arXiv preprint arXiv:1602.07360<\/u>, 2016 . Forrest N Iandola, Song Han, Matthew W Moskewicz, Khalid Ashraf, William J Dally, and Kurt Keutzer. Squeezenet: Alexnet-level accuracy with 50x fewer parameters and < 0.5 mb model size. <u>arXiv preprint arXiv:1602.07360<\/u>, 2016."},{"key":"e_1_3_2_1_26_1","volume-title":"et al. Mnn: A universal and efficient inference engine. <u>arXiv preprint arXiv:2002.12418<\/u>","author":"Jiang Xiaotang","year":"2020","unstructured":"Xiaotang Jiang , Huan Wang , Yiliu Chen , Ziqi Wu , Lichuan Wang , Bin Zou , Yafeng Yang , Zongyang Cui , Yu Cai , Tianhang Yu , et al. Mnn: A universal and efficient inference engine. <u>arXiv preprint arXiv:2002.12418<\/u> , 2020 . Xiaotang Jiang, Huan Wang, Yiliu Chen, Ziqi Wu, Lichuan Wang, Bin Zou, Yafeng Yang, Zongyang Cui, Yu Cai, Tianhang Yu, et al. Mnn: A universal and efficient inference engine. <u>arXiv preprint arXiv:2002.12418<\/u>, 2020."},{"key":"e_1_3_2_1_27_1","volume-title":"Layer-centric memory reuse and data migration for extreme-scale deep learning on many-core architectures. <u>ACM Transactions on Architecture and Code Optimization (TACO)<\/u>, 15(3):1--26","author":"Jin Hai","year":"2018","unstructured":"Hai Jin , Bo Liu , Wenbin Jiang , Yang Ma , Xuanhua Shi , Bingsheng He , and Shaofeng Zhao . Layer-centric memory reuse and data migration for extreme-scale deep learning on many-core architectures. <u>ACM Transactions on Architecture and Code Optimization (TACO)<\/u>, 15(3):1--26 , 2018 . Hai Jin, Bo Liu, Wenbin Jiang, Yang Ma, Xuanhua Shi, Bingsheng He, and Shaofeng Zhao. Layer-centric memory reuse and data migration for extreme-scale deep learning on many-core architectures. <u>ACM Transactions on Architecture and Code Optimization (TACO)<\/u>, 15(3):1--26, 2018."},{"key":"e_1_3_2_1_28_1","doi-asserted-by":"crossref","unstructured":"Tian Jin and Seokin Hong. Split-cnn: Splitting window-based operations in convolutional neural networks for memory system optimization. In <u>Proceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating Systems<\/u> pages 835--847 2019. Tian Jin and Seokin Hong. Split-cnn: Splitting window-based operations in convolutional neural networks for memory system optimization. In <u>Proceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating Systems<\/u> pages 835--847 2019.","DOI":"10.1145\/3297858.3304038"},{"key":"e_1_3_2_1_29_1","volume-title":"Keith Bonawitz, Zachary Charles, Graham Cormode, Rachel Cummings, et al. Advances and open problems in federated learning. <u>arXiv preprint arXiv:1912.04977<\/u>","author":"Kairouz Peter","year":"2019","unstructured":"Peter Kairouz , H Brendan McMahan , Brendan Avent , Aur\u00e9lien Bellet , Mehdi Bennis , Arjun Nitin Bhagoji , Keith Bonawitz, Zachary Charles, Graham Cormode, Rachel Cummings, et al. Advances and open problems in federated learning. <u>arXiv preprint arXiv:1912.04977<\/u> , 2019 . Peter Kairouz, H Brendan McMahan, Brendan Avent, Aur\u00e9lien Bellet, Mehdi Bennis, Arjun Nitin Bhagoji, Keith Bonawitz, Zachary Charles, Graham Cormode, Rachel Cummings, et al. Advances and open problems in federated learning. <u>arXiv preprint arXiv:1912.04977<\/u>, 2019."},{"key":"e_1_3_2_1_30_1","volume-title":"Smartlmk: A memory reclamation scheme for improving user-perceived app launch time. <u>ACM Transactions on Embedded Computing Systems (TECS)<\/u>, 15(3):1--25","author":"Kim Sang-Hoon","year":"2016","unstructured":"Sang-Hoon Kim , Jinkyu Jeong , Jin-Soo Kim , and Seungryoul Maeng . Smartlmk: A memory reclamation scheme for improving user-perceived app launch time. <u>ACM Transactions on Embedded Computing Systems (TECS)<\/u>, 15(3):1--25 , 2016 . Sang-Hoon Kim, Jinkyu Jeong, Jin-Soo Kim, and Seungryoul Maeng. Smartlmk: A memory reclamation scheme for improving user-perceived app launch time. <u>ACM Transactions on Embedded Computing Systems (TECS)<\/u>, 15(3):1--25, 2016."},{"key":"e_1_3_2_1_31_1","first-page":"1","volume-title":"Low latency on-device inference using cooperative single-layer acceleration and processor-friendly quantization. In <u>Proceedings of the Fourteenth EuroSys Conference 2019<\/u>","author":"Kim Youngsok","year":"2019","unstructured":"Youngsok Kim , Joonsung Kim , Dongju Chae , Daehyun Kim , and Jangwoo Kim . μlayer : Low latency on-device inference using cooperative single-layer acceleration and processor-friendly quantization. In <u>Proceedings of the Fourteenth EuroSys Conference 2019<\/u> , pages 1 -- 15 , 2019 . Youngsok Kim, Joonsung Kim, Dongju Chae, Daehyun Kim, and Jangwoo Kim. μlayer: Low latency on-device inference using cooperative single-layer acceleration and processor-friendly quantization. In <u>Proceedings of the Fourteenth EuroSys Conference 2019<\/u>, pages 1--15, 2019."},{"key":"e_1_3_2_1_32_1","first-page":"1","volume-title":"Spinn: synergistic progressive inference of neural networks over device and cloud. In <u>Proceedings of the 26th Annual International Conference on Mobile Computing and Networking<\/u>","author":"Laskaridis Stefanos","year":"2020","unstructured":"Stefanos Laskaridis , Stylianos I Venieris , Mario Almeida , Ilias Leontiadis , and Nicholas D Lane . Spinn: synergistic progressive inference of neural networks over device and cloud. In <u>Proceedings of the 26th Annual International Conference on Mobile Computing and Networking<\/u> , pages 1 -- 15 , 2020 . Stefanos Laskaridis, Stylianos I Venieris, Mario Almeida, Ilias Leontiadis, and Nicholas D Lane. Spinn: synergistic progressive inference of neural networks over device and cloud. In <u>Proceedings of the 26th Annual International Conference on Mobile Computing and Networking<\/u>, pages 1--15, 2020."},{"key":"e_1_3_2_1_33_1","volume-title":"Tflms: Large model support in tensorflow by graph rewriting. <u>arXiv preprint arXiv:1807.02037<\/u>","author":"Le Tung D","year":"2018","unstructured":"Tung D Le , Haruki Imai , Yasushi Negishi , and Kiyokuni Kawachiya . Tflms: Large model support in tensorflow by graph rewriting. <u>arXiv preprint arXiv:1807.02037<\/u> , 2018 . Tung D Le, Haruki Imai, Yasushi Negishi, and Kiyokuni Kawachiya. Tflms: Large model support in tensorflow by graph rewriting. <u>arXiv preprint arXiv:1807.02037<\/u>, 2018."},{"key":"e_1_3_2_1_34_1","first-page":"873","volume-title":"End the senseless killing: Improving memory management for mobile operating systems. In <u>2020 {USENIX} Annual Technical Conference ({USENIX} {ATC} 20)<\/u>","author":"Lebeck Niel","year":"2020","unstructured":"Niel Lebeck , Arvind Krishnamurthy , Henry M Levy , and Irene Zhang . End the senseless killing: Improving memory management for mobile operating systems. In <u>2020 {USENIX} Annual Technical Conference ({USENIX} {ATC} 20)<\/u> , pages 873 -- 887 , 2020 . Niel Lebeck, Arvind Krishnamurthy, Henry M Levy, and Irene Zhang. End the senseless killing: Improving memory management for mobile operating systems. In <u>2020 {USENIX} Annual Technical Conference ({USENIX} {ATC} 20)<\/u>, pages 873--887, 2020."},{"key":"e_1_3_2_1_35_1","doi-asserted-by":"crossref","unstructured":"Joohyun Lee Kyunghan Lee Euijin Jeong Jaemin Jo and Ness B Shroff. Context-aware application scheduling in mobile systems: What will users do and not do next? In <u>Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing<\/u> pages 1235--1246 2016. Joohyun Lee Kyunghan Lee Euijin Jeong Jaemin Jo and Ness B Shroff. Context-aware application scheduling in mobile systems: What will users do and not do next? In <u>Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing<\/u> pages 1235--1246 2016.","DOI":"10.1145\/2971648.2971680"},{"key":"e_1_3_2_1_36_1","unstructured":"Seulki Lee and Shahriar Nirjon. Fast and scalable in-memory deep multitask learning via neural weight virtualization. In <u>Proceedings of the 18th International Conference on Mobile Systems Applications and Services<\/u> pages 175--190 2020. Seulki Lee and Shahriar Nirjon. Fast and scalable in-memory deep multitask learning via neural weight virtualization. In <u>Proceedings of the 18th International Conference on Mobile Systems Applications and Services<\/u> pages 175--190 2020."},{"key":"e_1_3_2_1_37_1","first-page":"1","volume-title":"Occlumency: Privacy-preserving remote deep-learning inference using sgx. In <u>The 25th Annual International Conference on Mobile Computing and Networking<\/u>","author":"Lee Taegyeong","year":"2019","unstructured":"Taegyeong Lee , Zhiqi Lin , Saumay Pushp , Caihua Li , Yunxin Liu , Youngki Lee , Fengyuan Xu , Chenren Xu , Lintao Zhang , and Junehwa Song . Occlumency: Privacy-preserving remote deep-learning inference using sgx. In <u>The 25th Annual International Conference on Mobile Computing and Networking<\/u> , pages 1 -- 17 , 2019 . Taegyeong Lee, Zhiqi Lin, Saumay Pushp, Caihua Li, Yunxin Liu, Youngki Lee, Fengyuan Xu, Chenren Xu, Lintao Zhang, and Junehwa Song. Occlumency: Privacy-preserving remote deep-learning inference using sgx. In <u>The 25th Annual International Conference on Mobile Computing and Networking<\/u>, pages 1--17, 2019."},{"key":"e_1_3_2_1_38_1","unstructured":"Sijin Li Zhi-Qiang Liu and Antoni B Chan. Heterogeneous multi-task learning for human pose estimation with deep convolutional neural network. In <u>Proceedings of the IEEE conference on computer vision and pattern recognition workshops<\/u> pages 482--489 2014. Sijin Li Zhi-Qiang Liu and Antoni B Chan. Heterogeneous multi-task learning for human pose estimation with deep convolutional neural network. In <u>Proceedings of the IEEE conference on computer vision and pattern recognition workshops<\/u> pages 482--489 2014."},{"key":"e_1_3_2_1_39_1","unstructured":"Jiaqi Ma Zhe Zhao Xinyang Yi Jilin Chen Lichan Hong and Ed H Chi. Modeling task relationships in multi-task learning with multi-gate mixture-of-experts. In <u>Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining<\/u> pages 1930--1939 2018. Jiaqi Ma Zhe Zhao Xinyang Yi Jilin Chen Lichan Hong and Ed H Chi. Modeling task relationships in multi-task learning with multi-gate mixture-of-experts. In <u>Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining<\/u> pages 1930--1939 2018."},{"key":"e_1_3_2_1_40_1","first-page":"7765","volume-title":"Packnet: Adding multiple tasks to a single network by iterative pruning. In <u>Proceedings of the IEEE conference on Computer Vision and Pattern Recognition<\/u>","author":"Mallya Arun","year":"2018","unstructured":"Arun Mallya and Svetlana Lazebnik . Packnet: Adding multiple tasks to a single network by iterative pruning. In <u>Proceedings of the IEEE conference on Computer Vision and Pattern Recognition<\/u> , pages 7765 -- 7773 , 2018 . Arun Mallya and Svetlana Lazebnik. Packnet: Adding multiple tasks to a single network by iterative pruning. In <u>Proceedings of the IEEE conference on Computer Vision and Pattern Recognition<\/u>, pages 7765--7773, 2018."},{"key":"e_1_3_2_1_41_1","volume-title":"PMLR","author":"McMahan Brendan","year":"2017","unstructured":"Brendan McMahan , Eider Moore , Daniel Ramage , Seth Hampson , and Blaise Aguera y Arcas . Communication-efficient learning of deep networks from decentralized data. In <u>Artificial Intelligence and Statistics<\/u>, pages 1273--1282 . PMLR , 2017 . Brendan McMahan, Eider Moore, Daniel Ramage, Seth Hampson, and Blaise Aguera y Arcas. Communication-efficient learning of deep networks from decentralized data. In <u>Artificial Intelligence and Statistics<\/u>, pages 1273--1282. PMLR, 2017."},{"key":"e_1_3_2_1_42_1","volume-title":"Training deeper models by gpu memory optimization on tensorflow. In <u>Proc. of ML Systems Workshop in NIPS<\/u>","author":"Meng Chen","year":"2017","unstructured":"Chen Meng , Minmin Sun , Jun Yang , Minghui Qiu , and Yang Gu . Training deeper models by gpu memory optimization on tensorflow. In <u>Proc. of ML Systems Workshop in NIPS<\/u> , 2017 . Chen Meng, Minmin Sun, Jun Yang, Minghui Qiu, and Yang Gu. Training deeper models by gpu memory optimization on tensorflow. In <u>Proc. of ML Systems Workshop in NIPS<\/u>, 2017."},{"key":"e_1_3_2_1_43_1","doi-asserted-by":"publisher","DOI":"10.1145\/3458864.3466628"},{"key":"e_1_3_2_1_44_1","unstructured":"Chaoyue Niu Fan Wu Shaojie Tang Lifeng Hua Rongfei Jia Chengfei Lv Zhihua Wu and Guihai Chen. Billion-scale federated learning on mobile clients: A submodel design with tunable privacy. In <u>Proceedings of the 26th Annual International Conference on Mobile Computing and Networking<\/u> pages 1--14 2020. Chaoyue Niu Fan Wu Shaojie Tang Lifeng Hua Rongfei Jia Chengfei Lv Zhihua Wu and Guihai Chen. Billion-scale federated learning on mobile clients: A submodel design with tunable privacy. In <u>Proceedings of the 26th Annual International Conference on Mobile Computing and Networking<\/u> pages 1--14 2020."},{"key":"e_1_3_2_1_45_1","volume-title":"et al. Pytorch: An imperative style, high-performance deep learning library. <u>arXiv preprint arXiv:1912.01703<\/u>","author":"Paszke Adam","year":"2019","unstructured":"Adam Paszke , Sam Gross , Francisco Massa , Adam Lerer , James Bradbury , Gregory Chanan , Trevor Killeen , Zeming Lin , Natalia Gimelshein , Luca Antiga , et al. Pytorch: An imperative style, high-performance deep learning library. <u>arXiv preprint arXiv:1912.01703<\/u> , 2019 . Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, et al. Pytorch: An imperative style, high-performance deep learning library. <u>arXiv preprint arXiv:1912.01703<\/u>, 2019."},{"key":"e_1_3_2_1_46_1","doi-asserted-by":"crossref","unstructured":"Xuan Peng Xuanhua Shi Hulin Dai Hai Jin Weiliang Ma Qian Xiong Fan Yang and Xuehai Qian. Capuchin: Tensor-based gpu memory management for deep learning. In <u>Proceedings of the Twenty-Fifth International Conference on Architectural Support for Programming Languages and Operating Systems<\/u> pages 891--905 2020. Xuan Peng Xuanhua Shi Hulin Dai Hai Jin Weiliang Ma Qian Xiong Fan Yang and Xuehai Qian. Capuchin: Tensor-based gpu memory management for deep learning. In <u>Proceedings of the Twenty-Fifth International Conference on Architectural Support for Programming Languages and Operating Systems<\/u> pages 891--905 2020.","DOI":"10.1145\/3373376.3378505"},{"key":"e_1_3_2_1_47_1","volume-title":"Memory-efficient implementation of densenets. <u>arXiv preprint arXiv:1707.06990<\/u>","author":"Pleiss Geoff","year":"2017","unstructured":"Geoff Pleiss , Danlu Chen , Gao Huang , Tongcheng Li , Laurens van der Maaten , and Kilian Q Weinberger . Memory-efficient implementation of densenets. <u>arXiv preprint arXiv:1707.06990<\/u> , 2017 . Geoff Pleiss, Danlu Chen, Gao Huang, Tongcheng Li, Laurens van der Maaten, and Kilian Q Weinberger. Memory-efficient implementation of densenets. <u>arXiv preprint arXiv:1707.06990<\/u>, 2017."},{"key":"e_1_3_2_1_48_1","volume-title":"Model compression via distillation and quantization. <u>arXiv preprint arXiv:1802.05668<\/u>","author":"Polino Antonio","year":"2018","unstructured":"Antonio Polino , Razvan Pascanu , and Dan Alistarh . Model compression via distillation and quantization. <u>arXiv preprint arXiv:1802.05668<\/u> , 2018 . Antonio Polino, Razvan Pascanu, and Dan Alistarh. Model compression via distillation and quantization. <u>arXiv preprint arXiv:1802.05668<\/u>, 2018."},{"key":"e_1_3_2_1_49_1","volume-title":"USENIX Association","author":"Qiao Aurick","year":"2021","unstructured":"Aurick Qiao , Sang Keun Choe , Suhas Jayaram Subramanya , Willie Neiswanger , Qirong Ho , Hao Zhang , Gregory R. Ganger , and Eric P. Xing . Pollux: Co-adaptive cluster scheduling for goodput-optimized deep learning. In <u>15th USENIX Symposium on Operating Systems Design and Implementation (OSDI 21)<\/u>, pages 1--18 . USENIX Association , July 2021 . Aurick Qiao, Sang Keun Choe, Suhas Jayaram Subramanya, Willie Neiswanger, Qirong Ho, Hao Zhang, Gregory R. Ganger, and Eric P. Xing. Pollux: Co-adaptive cluster scheduling for goodput-optimized deep learning. In <u>15th USENIX Symposium on Operating Systems Design and Implementation (OSDI 21)<\/u>, pages 1--18. USENIX Association, July 2021."},{"key":"e_1_3_2_1_50_1","volume-title":"Zero-infinity: Breaking the gpu memory wall for extreme scale deep learning. <u>arXiv preprint arXiv:2104.07857<\/u>","author":"Rajbhandari Samyam","year":"2021","unstructured":"Samyam Rajbhandari , Olatunji Ruwase , Jeff Rasley , Shaden Smith , and Yuxiong He . Zero-infinity: Breaking the gpu memory wall for extreme scale deep learning. <u>arXiv preprint arXiv:2104.07857<\/u> , 2021 . Samyam Rajbhandari, Olatunji Ruwase, Jeff Rasley, Shaden Smith, and Yuxiong He. Zero-infinity: Breaking the gpu memory wall for extreme scale deep learning. <u>arXiv preprint arXiv:2104.07857<\/u>, 2021."},{"key":"e_1_3_2_1_51_1","first-page":"525","volume-title":"Xnornet: Imagenet classification using binary convolutional neural networks. In <u>European conference on computer vision<\/u>","author":"Rastegari Mohammad","year":"2016","unstructured":"Mohammad Rastegari , Vicente Ordonez , Joseph Redmon , and Ali Farhadi . Xnornet: Imagenet classification using binary convolutional neural networks. In <u>European conference on computer vision<\/u> , pages 525 -- 542 . Springer , 2016 . Mohammad Rastegari, Vicente Ordonez, Joseph Redmon, and Ali Farhadi. Xnornet: Imagenet classification using binary convolutional neural networks. In <u>European conference on computer vision<\/u>, pages 525--542. Springer, 2016."},{"key":"e_1_3_2_1_52_1","first-page":"1","volume-title":"vdnn: Virtualized deep neural networks for scalable, memory-efficient neural network design. In <u>2016 49th Annual IEEE\/ACM International Symposium on Microarchitecture (MICRO)<\/u>","author":"Rhu Minsoo","year":"2016","unstructured":"Minsoo Rhu , Natalia Gimelshein , Jason Clemons , Arslan Zulfiqar , and Stephen W Keckler . vdnn: Virtualized deep neural networks for scalable, memory-efficient neural network design. In <u>2016 49th Annual IEEE\/ACM International Symposium on Microarchitecture (MICRO)<\/u> , pages 1 -- 13 . IEEE , 2016 . Minsoo Rhu, Natalia Gimelshein, Jason Clemons, Arslan Zulfiqar, and Stephen W Keckler. vdnn: Virtualized deep neural networks for scalable, memory-efficient neural network design. In <u>2016 49th Annual IEEE\/ACM International Symposium on Microarchitecture (MICRO)<\/u>, pages 1--13. IEEE, 2016."},{"key":"e_1_3_2_1_53_1","first-page":"4510","volume-title":"Mobilenetv2: Inverted residuals and linear bottlenecks. In <u>Proceedings of the IEEE conference on computer vision and pattern recognition<\/u>","author":"Sandler Mark","year":"2018","unstructured":"Mark Sandler , Andrew Howard , Menglong Zhu , Andrey Zhmoginov , and Liang-Chieh Chen . Mobilenetv2: Inverted residuals and linear bottlenecks. In <u>Proceedings of the IEEE conference on computer vision and pattern recognition<\/u> , pages 4510 -- 4520 , 2018 . Mark Sandler, Andrew Howard, Menglong Zhu, Andrey Zhmoginov, and Liang-Chieh Chen. Mobilenetv2: Inverted residuals and linear bottlenecks. In <u>Proceedings of the IEEE conference on computer vision and pattern recognition<\/u>, pages 4510--4520, 2018."},{"key":"e_1_3_2_1_54_1","volume-title":"Don't decay the learning rate, increase the batch size. <u>arXiv preprint arXiv:1711.00489<\/u>","author":"Smith Samuel L","year":"2017","unstructured":"Samuel L Smith , Pieter-Jan Kindermans , Chris Ying , and Quoc V Le . Don't decay the learning rate, increase the batch size. <u>arXiv preprint arXiv:1711.00489<\/u> , 2017 . Samuel L Smith, Pieter-Jan Kindermans, Chris Ying, and Quoc V Le. Don't decay the learning rate, increase the batch size. <u>arXiv preprint arXiv:1711.00489<\/u>, 2017."},{"key":"e_1_3_2_1_55_1","volume-title":"Megan Leszczynski, Jian Zhang, and Christopher R\u00e9. Low-memory neural network training: A technical report. <u>arXiv preprint arXiv:1904.10631<\/u>","author":"Sohoni Nimit Sharad","year":"2019","unstructured":"Nimit Sharad Sohoni , Christopher Richard Aberger , Megan Leszczynski, Jian Zhang, and Christopher R\u00e9. Low-memory neural network training: A technical report. <u>arXiv preprint arXiv:1904.10631<\/u> , 2019 . Nimit Sharad Sohoni, Christopher Richard Aberger, Megan Leszczynski, Jian Zhang, and Christopher R\u00e9. Low-memory neural network training: A technical report. <u>arXiv preprint arXiv:1904.10631<\/u>, 2019."},{"key":"e_1_3_2_1_56_1","unstructured":"Sam Son Seung Yul Lee Yunho Jin Jonghyun Bae Jinkyu Jeong Tae Jun Ham Jae W Lee and Hongil Yoon. {ASAP}: Fast mobile application switch via adaptive prepaging. In <u>2021 {USENIX} Annual Technical Conference ({USENIX}{ATC } 21<\/u>) pages 365--380 2021. Sam Son Seung Yul Lee Yunho Jin Jonghyun Bae Jinkyu Jeong Tae Jun Ham Jae W Lee and Hongil Yoon. {ASAP}: Fast mobile application switch via adaptive prepaging. In <u>2021 {USENIX} Annual Technical Conference ({USENIX}{ATC } 21<\/u>) pages 365--380 2021."},{"key":"e_1_3_2_1_57_1","volume-title":"An information theoretic approach for combining neural network process models. <u>Neural Networks<\/u>, 12(6):915--926","author":"Sridhar Dasaratha V","year":"1999","unstructured":"Dasaratha V Sridhar , Eric B Bartlett , and Richard C Seagrave . An information theoretic approach for combining neural network process models. <u>Neural Networks<\/u>, 12(6):915--926 , 1999 . Dasaratha V Sridhar, Eric B Bartlett, and Richard C Seagrave. An information theoretic approach for combining neural network process models. <u>Neural Networks<\/u>, 12(6):915--926, 1999."},{"key":"e_1_3_2_1_58_1","volume-title":"NIPS 2011<\/u>","author":"Vanhoucke Vincent","year":"2011","unstructured":"Vincent Vanhoucke , Andrew Senior , and Mark Z Mao . Improving the speed of neural networks on cpus. In <u>Deep Learning and Unsupervised Feature Learning Workshop , NIPS 2011<\/u> , 2011 . Vincent Vanhoucke, Andrew Senior, and Mark Z Mao. Improving the speed of neural networks on cpus. In <u>Deep Learning and Unsupervised Feature Learning Workshop, NIPS 2011<\/u>, 2011."},{"key":"e_1_3_2_1_59_1","first-page":"5998","volume-title":"Attention is all you need. In <u>Advances in neural information processing systems<\/u>","author":"Vaswani Ashish","year":"2017","unstructured":"Ashish Vaswani , Noam Shazeer , Niki Parmar , Jakob Uszkoreit , Llion Jones , Aidan N Gomez , \u0141ukasz Kaiser , and Illia Polosukhin . Attention is all you need. In <u>Advances in neural information processing systems<\/u> , pages 5998 -- 6008 , 2017 . Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, \u0141ukasz Kaiser, and Illia Polosukhin. Attention is all you need. In <u>Advances in neural information processing systems<\/u>, pages 5998--6008, 2017."},{"key":"e_1_3_2_1_60_1","volume-title":"Split learning for health: Distributed deep learning without sharing raw patient data. <u>arXiv preprint arXiv:1812.00564<\/u>","author":"Vepakomma Praneeth","year":"2018","unstructured":"Praneeth Vepakomma , Otkrist Gupta , Tristan Swedish , and Ramesh Raskar . Split learning for health: Distributed deep learning without sharing raw patient data. <u>arXiv preprint arXiv:1812.00564<\/u> , 2018 . Praneeth Vepakomma, Otkrist Gupta, Tristan Swedish, and Ramesh Raskar. Split learning for health: Distributed deep learning without sharing raw patient data. <u>arXiv preprint arXiv:1812.00564<\/u>, 2018."},{"key":"e_1_3_2_1_61_1","doi-asserted-by":"crossref","unstructured":"Linnan Wang Jinmian Ye Yiyang Zhao Wei Wu Ang Li Shuaiwen Leon Song Zenglin Xu and Tim Kraska. Superneurons: Dynamic gpu memory management for training deep neural networks. In <u>Proceedings of the 23rd ACM SIGPLAN symposium on principles and practice of parallel programming<\/u> pages 41--53 2018. Linnan Wang Jinmian Ye Yiyang Zhao Wei Wu Ang Li Shuaiwen Leon Song Zenglin Xu and Tim Kraska. Superneurons: Dynamic gpu memory management for training deep neural networks. In <u>Proceedings of the 23rd ACM SIGPLAN symposium on principles and practice of parallel programming<\/u> pages 41--53 2018.","DOI":"10.1145\/3200691.3178491"},{"key":"e_1_3_2_1_62_1","first-page":"215","volume-title":"Asymo: scalable and efficient deep-learning inference on asymmetric mobile cpus. In <u>MobiCom<\/u>","author":"Wang Manni","year":"2021","unstructured":"Manni Wang , Shaohua Ding , Ting Cao , Yunxin Liu , and Fengyuan Xu . Asymo: scalable and efficient deep-learning inference on asymmetric mobile cpus. In <u>MobiCom<\/u> , pages 215 -- 228 , 2021 . Manni Wang, Shaohua Ding, Ting Cao, Yunxin Liu, and Fengyuan Xu. Asymo: scalable and efficient deep-learning inference on asymmetric mobile cpus. In <u>MobiCom<\/u>, pages 215--228, 2021."},{"key":"e_1_3_2_1_63_1","volume-title":"Philip HW Leong, and Hayden KH So. Niti: Training integer neural networks using integer-only arithmetic. <u>arXiv preprint arXiv:2009.13108<\/u>","author":"Wang Maolin","year":"2020","unstructured":"Maolin Wang , Seyedramin Rasoulinezhad , Philip HW Leong, and Hayden KH So. Niti: Training integer neural networks using integer-only arithmetic. <u>arXiv preprint arXiv:2009.13108<\/u> , 2020 . Maolin Wang, Seyedramin Rasoulinezhad, Philip HW Leong, and Hayden KH So. Niti: Training integer neural networks using integer-only arithmetic. <u>arXiv preprint arXiv:2009.13108<\/u>, 2020."},{"key":"e_1_3_2_1_64_1","volume-title":"Statistical theory of quantization. <u&gt","author":"Widrow Bernard","year":"1996","unstructured":"Bernard Widrow , Istvan Kollar , and Ming-Chang Liu . Statistical theory of quantization. <u&gt ; IEEE Transactions on instrumentation and measurement<\/u>, 45(2):353--361, 1996 . Bernard Widrow, Istvan Kollar, and Ming-Chang Liu. Statistical theory of quantization. <u>IEEE Transactions on instrumentation and measurement<\/u>, 45(2):353--361, 1996."},{"key":"e_1_3_2_1_65_1","first-page":"37","volume-title":"From cloud to edge: a first look at public edge platforms. In <u>Proceedings of the 21st ACM Internet Measurement Conference<\/u>","author":"Xu Mengwei","year":"2021","unstructured":"Mengwei Xu , Zhe Fu , Xiao Ma , Li Zhang , Yanan Li , Feng Qian , Shangguang Wang , Ke Li , Jingyu Yang , and Xuanzhe Liu . From cloud to edge: a first look at public edge platforms. In <u>Proceedings of the 21st ACM Internet Measurement Conference<\/u> , pages 37 -- 53 , 2021 . Mengwei Xu, Zhe Fu, Xiao Ma, Li Zhang, Yanan Li, Feng Qian, Shangguang Wang, Ke Li, Jingyu Yang, and Xuanzhe Liu. From cloud to edge: a first look at public edge platforms. In <u>Proceedings of the 21st ACM Internet Measurement Conference<\/u>, pages 37--53, 2021."},{"key":"e_1_3_2_1_66_1","first-page":"2125","volume-title":"Yunxin Liu, and Xuanzhe Liu. A first look at deep learning apps on smartphones. In <u>The World Wide Web Conference<\/u>","author":"Xu Mengwei","year":"2019","unstructured":"Mengwei Xu , Jiawei Liu , Yuanqiang Liu , Felix Xiaozhu Lin , Yunxin Liu, and Xuanzhe Liu. A first look at deep learning apps on smartphones. In <u>The World Wide Web Conference<\/u> , pages 2125 -- 2136 , 2019 . Mengwei Xu, Jiawei Liu, Yuanqiang Liu, Felix Xiaozhu Lin, Yunxin Liu, and Xuanzhe Liu. A first look at deep learning apps on smartphones. In <u>The World Wide Web Conference<\/u>, pages 2125--2136, 2019."},{"key":"e_1_3_2_1_67_1","volume-title":"Deeptype: On-device deep learning for input personalization service with minimal privacy concern. <u>IMWUT<\/u>, 2(4):1--26","author":"Xu Mengwei","year":"2018","unstructured":"Mengwei Xu , Feng Qian , Qiaozhu Mei , Kang Huang , and Xuanzhe Liu . Deeptype: On-device deep learning for input personalization service with minimal privacy concern. <u>IMWUT<\/u>, 2(4):1--26 , 2018 . Mengwei Xu, Feng Qian, Qiaozhu Mei, Kang Huang, and Xuanzhe Liu. Deeptype: On-device deep learning for input personalization service with minimal privacy concern. <u>IMWUT<\/u>, 2(4):1--26, 2018."},{"key":"e_1_3_2_1_68_1","volume-title":"Deepwear: Adaptive local offloading for on-wearable deep learning. <u&gt","author":"Xu Mengwei","year":"2019","unstructured":"Mengwei Xu , Feng Qian , Mengze Zhu , Feifan Huang , Saumay Pushp , and Xuanzhe Liu . Deepwear: Adaptive local offloading for on-wearable deep learning. <u&gt ;IEEE Transactions on Mobile Computing <\/u>, 19(2):314--330, 2019 . Mengwei Xu, Feng Qian, Mengze Zhu, Feifan Huang, Saumay Pushp, and Xuanzhe Liu. Deepwear: Adaptive local offloading for on-wearable deep learning. <u>IEEE Transactions on Mobile Computing<\/u>, 19(2):314--330, 2019."},{"key":"e_1_3_2_1_69_1","first-page":"191","volume-title":"Approximate query service on autonomous iot cameras. In <u>Proceedings of the 18th International Conference on Mobile Systems, Applications, and Services<\/u>","author":"Xu Mengwei","year":"2020","unstructured":"Mengwei Xu , Xiwen Zhang , Yunxin Liu , Gang Huang , Xuanzhe Liu , and Felix Xiaozhu Lin . Approximate query service on autonomous iot cameras. In <u>Proceedings of the 18th International Conference on Mobile Systems, Applications, and Services<\/u> , pages 191 -- 205 , 2020 . Mengwei Xu, Xiwen Zhang, Yunxin Liu, Gang Huang, Xuanzhe Liu, and Felix Xiaozhu Lin. Approximate query service on autonomous iot cameras. In <u>Proceedings of the 18th International Conference on Mobile Systems, Applications, and Services<\/u>, pages 191--205, 2020."},{"key":"e_1_3_2_1_70_1","first-page":"129","volume-title":"Felix Xiaozhu Lin, and Xuanzhe Liu. Deepcache: Principled cache for mobile deep vision. In <u>Proceedings of the 24th Annual International Conference on Mobile Computing and Networking<\/u>","author":"Xu Mengwei","year":"2018","unstructured":"Mengwei Xu , Mengze Zhu , Yunxin Liu , Felix Xiaozhu Lin, and Xuanzhe Liu. Deepcache: Principled cache for mobile deep vision. In <u>Proceedings of the 24th Annual International Conference on Mobile Computing and Networking<\/u> , pages 129 -- 144 , 2018 . Mengwei Xu, Mengze Zhu, Yunxin Liu, Felix Xiaozhu Lin, and Xuanzhe Liu. Deepcache: Principled cache for mobile deep vision. In <u>Proceedings of the 24th Annual International Conference on Mobile Computing and Networking<\/u>, pages 129--144, 2018."},{"key":"e_1_3_2_1_71_1","first-page":"113","volume-title":"Fast app launching for mobile devices using predictive user context. In <u>Proceedings of the 10th international conference on Mobile systems, applications, and services<\/u>","author":"Yan Tingxin","year":"2012","unstructured":"Tingxin Yan , David Chu , Deepak Ganesan , Aman Kansal , and Jie Liu . Fast app launching for mobile devices using predictive user context. In <u>Proceedings of the 10th international conference on Mobile systems, applications, and services<\/u> , pages 113 -- 126 , 2012 . Tingxin Yan, David Chu, Deepak Ganesan, Aman Kansal, and Jie Liu. Fast app launching for mobile devices using predictive user context. In <u>Proceedings of the 10th international conference on Mobile systems, applications, and services<\/u>, pages 113--126, 2012."},{"key":"e_1_3_2_1_72_1","volume-title":"Heterogeneity-aware federated learning. <u>arXiv preprint arXiv:2006.06983<\/u>","author":"Yang Chengxu","year":"2020","unstructured":"Chengxu Yang , QiPeng Wang , Mengwei Xu , Shangguang Wang , Kaigui Bian , and Xuanzhe Liu . Heterogeneity-aware federated learning. <u>arXiv preprint arXiv:2006.06983<\/u> , 2020 . Chengxu Yang, QiPeng Wang, Mengwei Xu, Shangguang Wang, Kaigui Bian, and Xuanzhe Liu. Heterogeneity-aware federated learning. <u>arXiv preprint arXiv:2006.06983<\/u>, 2020."},{"key":"e_1_3_2_1_73_1","unstructured":"Hyunho Yeo Chan Ju Chong Youngmok Jung Juncheol Ye and Dongsu Han. Nemo: enabling neural-enhanced video streaming on commodity mobile devices. In <u>Proceedings of the 26th Annual International Conference on Mobile Computing and Networking<\/u> pages 1--14 2020. Hyunho Yeo Chan Ju Chong Youngmok Jung Juncheol Ye and Dongsu Han. Nemo: enabling neural-enhanced video streaming on commodity mobile devices. In <u>Proceedings of the 26th Annual International Conference on Mobile Computing and Networking<\/u> pages 1--14 2020."},{"key":"e_1_3_2_1_74_1","first-page":"1","volume-title":"Dynamic control flow in large-scale machine learning. In <u>Proceedings of the Thirteenth EuroSys Conference<\/u>","author":"Yu Yuan","year":"2018","unstructured":"Yuan Yu , Mart\u00edn Abadi , Paul Barham , Eugene Brevdo , Mike Burrows , Andy Davis , Jeff Dean , Sanjay Ghemawat , Tim Harley , Peter Hawkins , Dynamic control flow in large-scale machine learning. In <u>Proceedings of the Thirteenth EuroSys Conference<\/u> , pages 1 -- 15 , 2018 . Yuan Yu, Mart\u00edn Abadi, Paul Barham, Eugene Brevdo, Mike Burrows, Andy Davis, Jeff Dean, Sanjay Ghemawat, Tim Harley, Peter Hawkins, et al. Dynamic control flow in large-scale machine learning. In <u>Proceedings of the Thirteenth EuroSys Conference<\/u>, pages 1--15, 2018."},{"key":"e_1_3_2_1_75_1","volume-title":"Yao Shu, Bingsheng He, and Wei Wang. Efficient memory management for gpu-based deep learning systems. <u>arXiv preprint arXiv:1903.06631<\/u>","author":"Zhang Junzhe","year":"2019","unstructured":"Junzhe Zhang , Sai Ho Yeung , Yao Shu, Bingsheng He, and Wei Wang. Efficient memory management for gpu-based deep learning systems. <u>arXiv preprint arXiv:1903.06631<\/u> , 2019 . Junzhe Zhang, Sai Ho Yeung, Yao Shu, Bingsheng He, and Wei Wang. Efficient memory management for gpu-based deep learning systems. <u>arXiv preprint arXiv:1903.06631<\/u>, 2019."},{"key":"e_1_3_2_1_76_1","first-page":"3298","volume-title":"A comprehensive benchmark of deep learning libraries on mobile devices. In <u>Proceedings of the ACM Web Conference 2022<\/u>","author":"Zhang Qiyang","year":"2022","unstructured":"Qiyang Zhang , Xiang Li , Xiangying Che , Xiao Ma , Ao Zhou , Mengwei Xu , Shangguang Wang , Yun Ma , and Xuanzhe Liu . A comprehensive benchmark of deep learning libraries on mobile devices. In <u>Proceedings of the ACM Web Conference 2022<\/u> , pages 3298 -- 3307 , 2022 . Qiyang Zhang, Xiang Li, Xiangying Che, Xiao Ma, Ao Zhou, Mengwei Xu, Shangguang Wang, Yun Ma, and Xuanzhe Liu. A comprehensive benchmark of deep learning libraries on mobile devices. In <u>Proceedings of the ACM Web Conference 2022<\/u>, pages 3298--3307, 2022."},{"key":"e_1_3_2_1_77_1","volume-title":"Dynamic memory optimization using pool allocation and prefetching. <u>ACM SIGARCH Computer Architecture News<\/u>, 33(5):27--32","author":"Zhao Qin","year":"2005","unstructured":"Qin Zhao , Rodric Rabbah , and Weng-Fai Wong . Dynamic memory optimization using pool allocation and prefetching. <u>ACM SIGARCH Computer Architecture News<\/u>, 33(5):27--32 , 2005 . Qin Zhao, Rodric Rabbah, and Weng-Fai Wong. Dynamic memory optimization using pool allocation and prefetching. <u>ACM SIGARCH Computer Architecture News<\/u>, 33(5):27--32, 2005."},{"key":"e_1_3_2_1_78_1","first-page":"177","volume-title":"Octo: Int8 training with loss-aware compensation and backward quantization for tiny on-device learning. In <u>2021 {USENIX} Annual Technical Conference ({USENIX} {ATC} 21)<\/u>","author":"Zhou Qihua","year":"2021","unstructured":"Qihua Zhou , Song Guo , Zhihao Qu , Jingcai Guo , Zhenda Xu , Jiewei Zhang , Tao Guo , Boyuan Luo , and Jingren Zhou . Octo: Int8 training with loss-aware compensation and backward quantization for tiny on-device learning. In <u>2021 {USENIX} Annual Technical Conference ({USENIX} {ATC} 21)<\/u> , pages 177 -- 191 , 2021 . Qihua Zhou, Song Guo, Zhihao Qu, Jingcai Guo, Zhenda Xu, Jiewei Zhang, Tao Guo, Boyuan Luo, and Jingren Zhou. Octo: Int8 training with loss-aware compensation and backward quantization for tiny on-device learning. In <u>2021 {USENIX} Annual Technical Conference ({USENIX} {ATC} 21)<\/u>, pages 177--191, 2021."}],"event":{"name":"MobiSys '22: The 20th Annual International Conference on Mobile Systems, Applications and Services","location":"Portland Oregon","acronym":"MobiSys '22","sponsor":["SIGMOBILE ACM Special Interest Group on Mobility of Systems, Users, Data and Computing","SIGOPS ACM Special Interest Group on Operating Systems"]},"container-title":["Proceedings of the 20th Annual International Conference on Mobile Systems, Applications and Services"],"original-title":[],"link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3498361.3538928","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3498361.3538928","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T18:10:04Z","timestamp":1750183804000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3498361.3538928"}},"subtitle":["breaking the memory wall for resource-efficient on-device machine learning"],"short-title":[],"issued":{"date-parts":[[2022,6,27]]},"references-count":78,"alternative-id":["10.1145\/3498361.3538928","10.1145\/3498361"],"URL":"https:\/\/doi.org\/10.1145\/3498361.3538928","relation":{},"subject":[],"published":{"date-parts":[[2022,6,27]]},"assertion":[{"value":"2022-06-27","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}