{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,2,21]],"date-time":"2025-02-21T14:43:43Z","timestamp":1740149023825,"version":"3.37.3"},"reference-count":22,"publisher":"Wiley","license":[{"start":{"date-parts":[[2018,5,14]],"date-time":"2018-05-14T00:00:00Z","timestamp":1526256000000},"content-version":"unspecified","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Hungarian Government","award":["GINOP-2.3.2-15-2016-00037"],"award-info":[{"award-number":["GINOP-2.3.2-15-2016-00037"]}]},{"DOI":"10.13039\/501100008530","name":"European Regional Development Fund","doi-asserted-by":"publisher","award":["GINOP-2.3.2-15-2016-00037"],"award-info":[{"award-number":["GINOP-2.3.2-15-2016-00037"]}],"id":[{"id":"10.13039\/501100008530","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Security and Communication Networks"],"published-print":{"date-parts":[[2018,5,14]]},"abstract":"<jats:p>Privacy and security are among the highest priorities in data mining approaches over data collected from mobile devices. Fully distributed machine learning is a promising direction in this context. However, it is a hard problem to design protocols that are efficient yet provide sufficient levels of privacy and security. In fully distributed environments, secure multiparty computation (MPC) is often applied to solve these problems. However, in our dynamic and unreliable application domain, known MPC algorithms are not scalable or not robust enough. We propose a light-weight protocol to quickly and securely compute the sum query over a subset of participants assuming a semihonest adversary. During the computation the participants learn no individual values. We apply this protocol to efficiently calculate the sum of gradients as part of a fully distributed minibatch stochastic gradient descent algorithm. The protocol achieves scalability and robustness by exploiting the fact that in this application domain a \u201cquick and dirty\u201d sum computation is acceptable. We utilize the Paillier homomorphic cryptosystem as part of our solution combined with extreme lossy gradient compression to make the cost of the cryptographic algorithms affordable. We demonstrate both theoretically and experimentally, based on churn statistics from a real smartphone trace, that the protocol is indeed practically viable.<\/jats:p>","DOI":"10.1155\/2018\/6728020","type":"journal-article","created":{"date-parts":[[2018,5,14]],"date-time":"2018-05-14T19:34:35Z","timestamp":1526326475000},"page":"1-15","source":"Crossref","is-referenced-by-count":5,"title":["Robust Fully Distributed Minibatch Gradient Descent with Privacy Preservation"],"prefix":"10.1155","volume":"2018","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9983-1060","authenticated-orcid":true,"given":"G\u00e1bor","family":"Danner","sequence":"first","affiliation":[{"name":"University of Szeged, and MTA-SZTE Research Group on AI, Szeged, Hungary"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4005-2273","authenticated-orcid":true,"given":"\u00c1rp\u00e1d","family":"Berta","sequence":"additional","affiliation":[{"name":"University of Szeged, and MTA-SZTE Research Group on AI, Szeged, Hungary"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5356-2192","authenticated-orcid":true,"given":"Istv\u00e1n","family":"Heged\u0171s","sequence":"additional","affiliation":[{"name":"University of Szeged, and MTA-SZTE Research Group on AI, Szeged, Hungary"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9363-1482","authenticated-orcid":true,"given":"M\u00e1rk","family":"Jelasity","sequence":"additional","affiliation":[{"name":"University of Szeged, and MTA-SZTE Research Group on AI, Szeged, Hungary"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"311","reference":[{"key":"16","doi-asserted-by":"publisher","DOI":"10.1109\/MLSP.2016.7738826"},{"key":"23","doi-asserted-by":"publisher","DOI":"10.1002\/cpe.2858"},{"key":"15","doi-asserted-by":"publisher","DOI":"10.1016\/j.comnet.2010.03.020"},{"key":"10","first-page":"165","volume":"13","year":"2012","journal-title":"Journal of Machine Learning Research (JMLR)"},{"first-page":"160","volume-title":"Protocols for secure computations","year":"1982","key":"31"},{"key":"8","doi-asserted-by":"publisher","DOI":"10.1145\/772862.772867"},{"key":"27","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-662-46078-8_3"},{"key":"3","doi-asserted-by":"publisher","DOI":"10.1007\/s12083-009-0051-9"},{"key":"22","doi-asserted-by":"publisher","DOI":"10.1007\/s00607-012-0200-5"},{"key":"14","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2009.153"},{"key":"11","doi-asserted-by":"publisher","DOI":"10.1145\/1866739.1866758"},{"key":"12","doi-asserted-by":"publisher","DOI":"10.1007\/11761679_29"},{"key":"9","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-19129-4_3"},{"key":"30","doi-asserted-by":"publisher","DOI":"10.1109\/TNET.2008.2001730"},{"key":"19","doi-asserted-by":"publisher","DOI":"10.1016\/j.dam.2005.03.020"},{"key":"6","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-35289-8_25"},{"volume-title":"Large scale online learning","year":"2004","key":"7"},{"key":"24","doi-asserted-by":"publisher","DOI":"10.1007\/3-540-48910-X_16"},{"year":"2013","key":"17"},{"year":"2006","key":"4"},{"key":"28","doi-asserted-by":"publisher","DOI":"10.1007\/s10107-010-0420-4"},{"year":"2017","key":"29"}],"container-title":["Security and Communication Networks"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/downloads.hindawi.com\/journals\/scn\/2018\/6728020.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/downloads.hindawi.com\/journals\/scn\/2018\/6728020.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/downloads.hindawi.com\/journals\/scn\/2018\/6728020.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2018,6,6]],"date-time":"2018-06-06T19:31:58Z","timestamp":1528313518000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.hindawi.com\/journals\/scn\/2018\/6728020\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2018,5,14]]},"references-count":22,"alternative-id":["6728020","6728020"],"URL":"https:\/\/doi.org\/10.1155\/2018\/6728020","relation":{},"ISSN":["1939-0114","1939-0122"],"issn-type":[{"type":"print","value":"1939-0114"},{"type":"electronic","value":"1939-0122"}],"subject":[],"published":{"date-parts":[[2018,5,14]]}}}