{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,19]],"date-time":"2025-12-19T09:59:35Z","timestamp":1766138375166,"version":"3.41.0"},"reference-count":49,"publisher":"Association for Computing Machinery (ACM)","issue":"3","license":[{"start":{"date-parts":[[2024,5,11]],"date-time":"2024-05-11T00:00:00Z","timestamp":1715385600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"ARO-MURI","award":["W911NF-21-1-0322"],"award-info":[{"award-number":["W911NF-21-1-0322"]}]},{"name":"NSF-CNS","award":["2016737"],"award-info":[{"award-number":["2016737"]}]},{"name":"Intel PrivateAI Collaborative Research Institute"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Embed. Comput. Syst."],"published-print":{"date-parts":[[2024,5,31]]},"abstract":"<jats:p>Binary neural network (BNN) delivers increased compute intensity and reduces memory\/data requirements for computation. Scalable BNN enables inference in a limited time due to different constraints. This paper explores the application of Scalable BNN in oblivious inference, a service provided by a server to mistrusting clients. Using this service, a client can obtain the inference result on his\/her data by a trained model held by the server without disclosing the data or learning the model parameters. Two contributions of this paper are: (1) we devise lightweight cryptographic protocols explicitly designed to exploit the unique characteristics of BNNs. (2) we present an advanced dynamic exploration of the runtime-accuracy tradeoff of scalable BNNs in a single-shot training process. While previous works trained multiple BNNs with different computational complexities (which is cumbersome due to the slow convergence of BNNs), we train a single BNN that can perform inference under various computational budgets. Compared to CryptFlow2, the state-of-the-art technique in the oblivious inference of non-binary DNNs, our approach reaches 3\u00d7 faster inference while keeping the same accuracy. Compared to XONN, the state-of-the-art technique in the oblivious inference of binary networks, we achieve 2\u00d7 to 12\u00d7 faster inference while obtaining higher accuracy.<\/jats:p>","DOI":"10.1145\/3607192","type":"journal-article","created":{"date-parts":[[2023,7,5]],"date-time":"2023-07-05T12:20:54Z","timestamp":1688559654000},"page":"1-18","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":4,"title":["Scalable Binary Neural Network Applications in Oblivious Inference"],"prefix":"10.1145","volume":"23","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2785-2321","authenticated-orcid":false,"given":"Xinqiao","family":"Zhang","sequence":"first","affiliation":[{"name":"UC San Diego and San Diego State University, La Jolla, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1404-527X","authenticated-orcid":false,"given":"Mohammad","family":"Samragh","sequence":"additional","affiliation":[{"name":"UC San Diego, La Jolla, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8668-0702","authenticated-orcid":false,"given":"Siam","family":"Hussain","sequence":"additional","affiliation":[{"name":"UC San Diego, La Jolla, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1587-9877","authenticated-orcid":false,"given":"Ke","family":"Huang","sequence":"additional","affiliation":[{"name":"San Diego State University, San Diego, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0798-3794","authenticated-orcid":false,"given":"Farinaz","family":"Koushanfar","sequence":"additional","affiliation":[{"name":"UC San Diego, La Jolla, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2024,5,11]]},"reference":[{"unstructured":"(n.d.). Malaria Cell Images accessed on 01\/20\/2019. https:\/\/www.kaggle.com\/iarunava\/cell-images-for-detecting-malaria.","key":"e_1_3_2_2_2"},{"doi-asserted-by":"publisher","key":"e_1_3_2_3_2","DOI":"10.1145\/2508859.2516738"},{"doi-asserted-by":"publisher","key":"e_1_3_2_4_2","DOI":"10.1145\/1029179.1029204"},{"key":"e_1_3_2_5_2","first-page":"338","article-title":"Garbled neural networks are practical","volume":"2019","author":"Ball Marshall","year":"2019","unstructured":"Marshall Ball, Brent Carmer, Tal Malkin, Mike Rosulek, and Nichole Schimanski. 2019. Garbled neural networks are practical. IACR Cryptol. ePrint Arch. 2019 (2019), 338.","journal-title":"IACR Cryptol. ePrint Arch."},{"key":"e_1_3_2_6_2","article-title":"MeliusNet: Can binary neural networks achieve mobilenet-level accuracy?","author":"Bethge Joseph","year":"2020","unstructured":"Joseph Bethge, Christian Bartz, Haojin Yang, Ying Chen, and Christoph Meinel. 2020. MeliusNet: Can binary neural networks achieve mobilenet-level accuracy? arXiv preprint arXiv:2001.05936 (2020).","journal-title":"arXiv preprint arXiv:2001.05936"},{"key":"e_1_3_2_7_2","first-page":"483","volume-title":"Annual International Cryptology Conference","author":"Bourse Florian","year":"2018","unstructured":"Florian Bourse, Michele Minelli, Matthias Minihold, and Pascal Paillier. 2018. Fast homomorphic evaluation of deep discretized neural networks. In Annual International Cryptology Conference. Springer, 483\u2013512."},{"key":"e_1_3_2_8_2","first-page":"812","volume-title":"International Conference on Machine Learning","author":"Brutzkus Alon","year":"2019","unstructured":"Alon Brutzkus, Ran Gilad-Bachrach, and Oren Elisha. 2019. Low latency privacy preserving inference. In International Conference on Machine Learning. PMLR, 812\u2013821."},{"key":"e_1_3_2_9_2","article-title":"EzPC: Programmable, efficient, and scalable secure two-party computation for machine learning","volume":"1109","author":"Chandran Nishanth","year":"2017","unstructured":"Nishanth Chandran, Divya Gupta, Aseem Rastogi, Rahul Sharma, and Shardul Tripathi. 2017. EzPC: Programmable, efficient, and scalable secure two-party computation for machine learning. ePrint Report 1109 (2017).","journal-title":"ePrint Report"},{"key":"e_1_3_2_10_2","article-title":"Faster CryptoNets: Leveraging sparsity for real-world encrypted inference","author":"Chou Edward","year":"2018","unstructured":"Edward Chou, Josh Beal, Daniel Levy, Serena Yeung, Albert Haque, and Li Fei-Fei. 2018. Faster CryptoNets: Leveraging sparsity for real-world encrypted inference. arXiv preprint arXiv:1811.09953 (2018).","journal-title":"arXiv preprint arXiv:1811.09953"},{"key":"e_1_3_2_11_2","first-page":"215","volume-title":"Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics","author":"Coates Adam","year":"2011","unstructured":"Adam Coates, Andrew Ng, and Honglak Lee. 2011. An analysis of single-layer networks in unsupervised feature learning. In Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics. JMLR Workshop and Conference Proceedings, 215\u2013223."},{"key":"e_1_3_2_12_2","article-title":"Binarized neural networks: Training deep neural networks with weights and activations constrained to+ 1 or-1","author":"Courbariaux Matthieu","year":"2016","unstructured":"Matthieu Courbariaux, Itay Hubara, Daniel Soudry, Ran El-Yaniv, and Yoshua Bengio. 2016. Binarized neural networks: Training deep neural networks with weights and activations constrained to+ 1 or-1. arXiv preprint arXiv:1602.02830 (2016).","journal-title":"arXiv preprint arXiv:1602.02830"},{"doi-asserted-by":"publisher","key":"e_1_3_2_13_2","DOI":"10.1145\/3314221.3314628"},{"doi-asserted-by":"publisher","key":"e_1_3_2_14_2","DOI":"10.1038\/s41591-018-0316-z"},{"volume-title":"The FaceScrub Dataset","year":"2020","unstructured":"FaceScrub. 2020. The FaceScrub Dataset. http:\/\/engineering.purdue.edu\/ mark\/puthesis, (accessed July 3, 2020).","key":"e_1_3_2_15_2"},{"key":"e_1_3_2_16_2","article-title":"The lottery ticket hypothesis: Finding sparse, trainable neural networks","author":"Frankle Jonathan","year":"2018","unstructured":"Jonathan Frankle and Michael Carbin. 2018. The lottery ticket hypothesis: Finding sparse, trainable neural networks. arXiv preprint arXiv:1803.03635 (2018).","journal-title":"arXiv preprint arXiv:1803.03635"},{"doi-asserted-by":"publisher","key":"e_1_3_2_17_2","DOI":"10.1109\/FCCM.2018.00018"},{"key":"e_1_3_2_18_2","volume-title":"Advances in Neural Information Processing Systems","author":"Ghodsi Zahra","year":"2020","unstructured":"Zahra Ghodsi, Akshaj Veldanda, Brandon Reagen, and Siddharth Garg. 2020. CryptoNAS: Private inference on a ReLU budget. In Advances in Neural Information Processing Systems."},{"key":"e_1_3_2_19_2","volume-title":"International Conference on Machine Learning","author":"Gilad-Bachrach Ran","year":"2016","unstructured":"Ran Gilad-Bachrach, Nathan Dowlin, Kim Laine, Kristin Lauter, Michael Naehrig, and John Wernsing. 2016. CryptoNets: Applying neural networks to encrypted data with high throughput and accuracy. In International Conference on Machine Learning."},{"doi-asserted-by":"publisher","key":"e_1_3_2_20_2","DOI":"10.1109\/ICCV.2017.155"},{"key":"e_1_3_2_21_2","article-title":"CryptoDL: Deep neural networks over encrypted data","author":"Hesamifard Ehsan","year":"2017","unstructured":"Ehsan Hesamifard, Hassan Takabi, and Mehdi Ghasemi. 2017. CryptoDL: Deep neural networks over encrypted data. arXiv preprint arXiv:1711.05189 (2017).","journal-title":"arXiv preprint arXiv:1711.05189"},{"issue":"7","key":"e_1_3_2_22_2","article-title":"Distilling the knowledge in a neural network","volume":"2","author":"Hinton Geoffrey","year":"2015","unstructured":"Geoffrey Hinton, Oriol Vinyals, Jeff Dean, et\u00a0al. 2015. Distilling the knowledge in a neural network. arXiv preprint arXiv:1503.02531 2, 7 (2015).","journal-title":"arXiv preprint arXiv:1503.02531"},{"doi-asserted-by":"publisher","key":"e_1_3_2_23_2","DOI":"10.1007\/978-3-540-45146-4_9"},{"key":"e_1_3_2_24_2","first-page":"1651","volume-title":"27th  \\(\\lbrace\\) USENIX \\(\\rbrace\\)  Security Symposium ( \\(\\lbrace\\) USENIX \\(\\rbrace\\)  Security 18)","author":"Juvekar Chiraag","year":"2018","unstructured":"Chiraag Juvekar, Vinod Vaikuntanathan, and Anantha Chandrakasan. 2018. \\(\\lbrace\\) GAZELLE \\(\\rbrace\\) : A low latency framework for secure neural network inference. In 27th \\(\\lbrace\\) USENIX \\(\\rbrace\\) Security Symposium ( \\(\\lbrace\\) USENIX \\(\\rbrace\\) Security 18). 1651\u20131669."},{"key":"e_1_3_2_25_2","article-title":"Compression of deep convolutional neural networks for fast and low power mobile applications","author":"Kim Yong-Deok","year":"2015","unstructured":"Yong-Deok Kim, Eunhyeok Park, Sungjoo Yoo, Taelim Choi, Lu Yang, and Dongjun Shin. 2015. Compression of deep convolutional neural networks for fast and low power mobile applications. arXiv preprint arXiv:1511.06530 (2015).","journal-title":"arXiv preprint arXiv:1511.06530"},{"doi-asserted-by":"publisher","key":"e_1_3_2_26_2","DOI":"10.1007\/978-3-540-70583-3_40"},{"key":"e_1_3_2_27_2","volume-title":"Electronic Colloquium on Computational Complexity (ECCC)","author":"Lindell Y.","year":"2004","unstructured":"Y. Lindell and B. Pinkas. 2004. A proof of Yao\u2019s protocol for secure two-party computation. ECCC Report TR04-063. In Electronic Colloquium on Computational Complexity (ECCC)."},{"doi-asserted-by":"publisher","key":"e_1_3_2_28_2","DOI":"10.1145\/3133956.3134056"},{"doi-asserted-by":"publisher","key":"e_1_3_2_29_2","DOI":"10.1609\/aaai.v32i1.11630"},{"doi-asserted-by":"publisher","key":"e_1_3_2_30_2","DOI":"10.1007\/978-3-030-58568-6_9"},{"key":"e_1_3_2_31_2","volume-title":"International Conference on Learning Representations","author":"Lou Qian","year":"2021","unstructured":"Qian Lou, Yilin Shen, Hongxia Jin, and Lei Jiang. 2021. {SAFEN}et: A secure, accurate and fast neural network inference. In International Conference on Learning Representations."},{"key":"e_1_3_2_32_2","volume-title":"Advances in Neural Information Processing Systems","author":"Lou Qian","year":"2020","unstructured":"Qian Lou, Bian Song, and Lei Jiang. 2020. AutoPrivacy: Automated layer-wise parameter selection for secure neural network inference. In Advances in Neural Information Processing Systems."},{"key":"e_1_3_2_33_2","article-title":"WRPN: Wide reduced-precision networks","author":"Mishra Asit","year":"2017","unstructured":"Asit Mishra, Eriko Nurvitadhi, Jeffrey J. Cook, and Debbie Marr. 2017. WRPN: Wide reduced-precision networks. arXiv preprint arXiv:1709.01134 (2017).","journal-title":"arXiv preprint arXiv:1709.01134"},{"key":"e_1_3_2_34_2","volume-title":"29th  \\(\\lbrace\\) USENIX \\(\\rbrace\\)  Security Symposium ( \\(\\lbrace\\) USENIX \\(\\rbrace\\)  Security 20)","author":"Mishra Pratyush","year":"2020","unstructured":"Pratyush Mishra, Ryan Lehmkuhl, Akshayaram Srinivasan, Wenting Zheng, and Raluca Ada Popa. 2020. DELPHI: A cryptographic inference service for neural networks. In 29th \\(\\lbrace\\) USENIX \\(\\rbrace\\) Security Symposium ( \\(\\lbrace\\) USENIX \\(\\rbrace\\) Security 20)."},{"key":"e_1_3_2_35_2","first-page":"112","volume-title":"EuroS&P","author":"Mood Benjamin","year":"2016","unstructured":"Benjamin Mood, Debayan Gupta, Henry Carter, Kevin Butler, and Patrick Traynor. 2016. Frigate: A validated, extensible, and efficient compiler and interpreter for secure computation. In EuroS&P. IEEE, 112\u2013127."},{"issue":"1","key":"e_1_3_2_36_2","article-title":"Computationally secure oblivious transfer","volume":"18","author":"Naor Moni","year":"2005","unstructured":"Moni Naor and Benny Pinkas. 2005. Computationally secure oblivious transfer. Journal of Cryptology 18, 1 (2005).","journal-title":"Journal of Cryptology"},{"key":"e_1_3_2_37_2","volume-title":"IEEE International Conference on Image Processing","author":"Ng Hong-Wei","year":"2014","unstructured":"Hong-Wei Ng and Stefan Winkler. 2014. A data-driven approach to cleaning large face datasets. In IEEE International Conference on Image Processing."},{"doi-asserted-by":"publisher","key":"e_1_3_2_38_2","DOI":"10.1145\/3372297.3417274"},{"key":"e_1_3_2_39_2","volume-title":"USENIX Security","author":"Riazi M. Sadegh","year":"2019","unstructured":"M. Sadegh Riazi, Mohammad Samragh, Hao Chen, Kim Laine, Kristin E. Lauter, and Farinaz Koushanfar. 2019. XONN: XNOR-based oblivious deep neural network inference. In USENIX Security."},{"doi-asserted-by":"publisher","key":"e_1_3_2_40_2","DOI":"10.1145\/3196494.3196522"},{"key":"e_1_3_2_41_2","article-title":"DeepSecure: Scalable provably-secure deep learning","author":"Rouhani Bita Darvish","year":"2017","unstructured":"Bita Darvish Rouhani, M. Sadegh Riazi, and Farinaz Koushanfar. 2017. DeepSecure: Scalable provably-secure deep learning. arXiv preprint arXiv:1705.08963 (2017).","journal-title":"arXiv preprint arXiv:1705.08963"},{"key":"e_1_3_2_42_2","first-page":"4490","volume-title":"International Conference on Machine Learning","author":"Sanyal Amartya","year":"2018","unstructured":"Amartya Sanyal, Matt Kusner, Adria Gascon, and Varun Kanade. 2018. TAPAS: Tricks to accelerate (encrypted) prediction as a service. In International Conference on Machine Learning. PMLR, 4490\u20134499."},{"key":"e_1_3_2_43_2","volume-title":"Proceedings of the AAAI Conference on Artificial Intelligence","volume":"31","author":"Tang Wei","year":"2017","unstructured":"Wei Tang, Gang Hua, and Liang Wang. 2017. How to train a compact binary neural network with high accuracy?. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 31."},{"unstructured":"The HIPAA Privacy Rule. (n.d.). https:\/\/www.hhs.gov\/hipaa\/for-professionals\/privacy\/index.html.","key":"e_1_3_2_44_2"},{"key":"e_1_3_2_45_2","article-title":"EMP-toolkit: Efficient MultiParty computation toolkit","author":"Wang Xiao","year":"2016","unstructured":"Xiao Wang, Alex J. Malozemoff, and Jonathan Katz. 2016. EMP-toolkit: Efficient MultiParty computation toolkit. https:\/\/github.com\/emp-toolkit. (2016).","journal-title":"https:\/\/github.com\/emp-toolkit"},{"unstructured":"Sophia Yakoubov. 2017. A gentle introduction to Yao\u2019s Garbled Circuits. (2017).","key":"e_1_3_2_46_2"},{"doi-asserted-by":"publisher","key":"e_1_3_2_47_2","DOI":"10.1109\/SFCS.1986.25"},{"doi-asserted-by":"publisher","key":"e_1_3_2_48_2","DOI":"10.1109\/ICCV.2019.00189"},{"key":"e_1_3_2_49_2","article-title":"Slimmable neural networks","author":"Yu Jiahui","year":"2018","unstructured":"Jiahui Yu, Linjie Yang, Ning Xu, Jianchao Yang, and Thomas Huang. 2018. Slimmable neural networks. arXiv preprint arXiv:1812.08928 (2018).","journal-title":"arXiv preprint arXiv:1812.08928"},{"key":"e_1_3_2_50_2","article-title":"DoReFa-Net: Training low bitwidth convolutional neural networks with low bitwidth gradients","author":"Zhou Shuchang","year":"2016","unstructured":"Shuchang Zhou, Yuxin Wu, Zekun Ni, Xinyu Zhou, He Wen, and Yuheng Zou. 2016. DoReFa-Net: Training low bitwidth convolutional neural networks with low bitwidth gradients. arXiv preprint arXiv:1606.06160 (2016).","journal-title":"arXiv preprint arXiv:1606.06160"}],"container-title":["ACM Transactions on Embedded Computing Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3607192","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3607192","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T16:37:34Z","timestamp":1750178254000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3607192"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,5,11]]},"references-count":49,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2024,5,31]]}},"alternative-id":["10.1145\/3607192"],"URL":"https:\/\/doi.org\/10.1145\/3607192","relation":{},"ISSN":["1539-9087","1558-3465"],"issn-type":[{"type":"print","value":"1539-9087"},{"type":"electronic","value":"1558-3465"}],"subject":[],"published":{"date-parts":[[2024,5,11]]},"assertion":[{"value":"2022-10-31","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2023-06-25","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2024-05-11","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}