{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/tapas-tricks-to-accelerate-encrypted","title":"TAPAS: Tricks to Accelerate (encrypted) Prediction As a Service","arxiv_id":"1806.03461","date":"2018-06-09","proceeding":"ICML 2018 7","authors":["Amartya Sanyal","Matt J. Kusner","Adrià Gascón","Varun Kanade"],"abstract":"Machine learning methods are widely used for a variety of prediction\nproblems. \\emph{Prediction as a service} is a paradigm in which service\nproviders with technological expertise and computational resources may perform\npredictions for clients. However, data privacy severely restricts the\napplicability of such services, unless measures to keep client data private\n(even from the service provider) are designed. Equally important is to minimize\nthe amount of computation and communication required between client and server.\nFully homomorphic encryption offers a possible way out, whereby clients may\nencrypt their data, and on which the server may perform arithmetic\ncomputations. The main drawback of using fully homomorphic encryption is the\namount of time required to evaluate large machine learning models on encrypted\ndata. We combine ideas from the machine learning literature, particularly work\non binarization and sparsification of neural networks, together with\nalgorithmic tools to speed-up and parallelize computation using encrypted data.","url_abs":"http://arxiv.org/abs/1806.03461v1","url_pdf":"http://arxiv.org/pdf/1806.03461v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"tapas-tricks-to-accelerate-encrypted","repo_url":"https://github.com/amartya18x/tapas","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"machine-learning","task_name":"BIG-bench Machine Learning"},{"task_slug":"binarization","task_name":"Binarization"},{"task_slug":"prediction","task_name":"Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1806.03461","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}