본문 바로가기
KERD
논문KCI 등재

프라이버시를 보호하는 분산 기계 학습 연구 동향

Systematic Research on Privacy-Preserving Distributed Machine Learning

신영아|천지영|이민섭
한국정보처리학회|2024.02|정보처리학회 논문지 (The Transactions of the Korea Information Processing Society)|제13권 제2호|pp. 76-90|DOI 10.3745/TKIPS.2024.13.2.76

1

조회

0

다운로드

0

인용

0

참고문헌

키워드

프라이버시 보호Federated LearningDistributed Machine Learning연합 학습분산 기계 학습스웜 학습Privacy-Preserving TechnologiesSwarm Learning
주제분류컴퓨터학

초록

인공지능 기술은 스마트 시티, 자율 주행, 의료 분야 등 다양한 분야에서 활용 가능성을 높이 평가받고 있으나, 정보주체의 개인정보 및 민감정보의노출 문제로 모델 활용이 제한되고 있다. 이에 따라 데이터를 중앙 서버에 모아서 학습하지 않고, 보유 데이터셋을 바탕으로 일차적으로 학습을진행한 후 글로벌 모델을 최종적으로 학습하는 분산 기계 학습의 개념이 등장하였다. 그러나, 분산 기계 학습은 여전히 협력하여 학습을 진행하는과정에서 데이터 프라이버시 위협이 발생한다. 본 연구는 분산 기계 학습 연구 분야에서 프라이버시를 보호하기 위한 연구를 서버의 존재 유무,학습 데이터셋의 분포 환경, 참여자의 성능 차이 등 현재까지 제안된 분류 기준들을 바탕으로 유기적으로 분석하여 최신 연구 동향을 파악한다. 특히, 대표적인 분산 기계 학습 기법인 수평적 연합학습, 수직적 연합학습, 스웜 학습에 집중하여 활용된 프라이버시 보호 기법을 살펴본 후 향후진행되어야 할 연구 방향을 모색한다.

Abstract

Although artificial intelligence (AI) can be utilized in various domains such as smart city, healthcare, it is limited due to concernsabout the exposure of personal and sensitive information. In response, the concept of distributed machine learning has emerged, whereinlearning occurs locally before training a global model, mitigating the concentration of data on a central server. However, overall learningphase in a collaborative way among multiple participants poses threats to data privacy. In this paper, we systematically analyzes recenttrends in privacy protection within the realm of distributed machine learning, considering factors such as the presence of a central server,distribution environment of the training datasets, and performance variations among participants. In particular, we focus on key distributedmachine learning techniques, including horizontal federated learning, vertical federated learning, and swarm learning. We examine privacyprotection mechanisms within these techniques and explores potential directions for future research.

참고문헌 (76)

  1. A. Shamir, "How to share a secret", Communications of the ACM, 22(11), pp.612~613 (1979)
  2. W. Diffie, "Democratizing Cryptography: The Work of Whitfield Diffie and Martin Hellman", pp.365~390 (2022)
  3. P. Paillier, "Public-key cryptosystems based on composite degree residuosity classes", International Conference on the Theory and Applications of Cryptographic Techniques, pp.223~238 (1999)
  4. Q. Li, "A survey on federated learning systems : Vision, hype and reality for data privacy and protection", IEEE Transactions on Knowledge and Data Engineering (2021)
  5. "What is Data Cleansing?"[URL]
  6. L. Ma, "Federated Data Cleaning : Collaborative and Privacy-Preserving Data Cleaning for Edge Intelligence", IEEE Internet of Things Journal, 8(8), pp.6757~6770 (2021)[DOI]
  7. A. Koufakou, "A scalable and efficient outlier detection strategy for categorical data", 19th IEEE International Conference on Tools with Artificial Intelligence (ICTAI), 2, pp.210~217 (2007)
  8. S. D. Bay, "Mining distance-based outliers in near linear time with randomization and a simple pruning rule", Proceedings of the Ninth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp.29~38 (2003)
  9. F. Jiang, "Initialization of K-modes clustering using outlier detection techniques", Information Sciences, 332, pp.167~183 (2016)
  10. M. M. Breunig, "LOF : Identifying density-based local outliers", Proceedings of the 2000 ACM SIGMOD international conference on Management of data, 29(2), pp.93~104 (2000)
  11. A. Arasu, "On active learning of record matching packages", Proceedings of the 2010ACM SIGMOD International Conference on Management of data, pp.783~794 (2010)
  12. S. Mudgal, "Deep learning for entity matching : A design space exploration", Proceedings of the 2018International Conference on Management of Data, 19(34), pp.19~34 (2018)
  13. T. Rekatsinas, "Holoclean : Holistic data repairs with probabilistic inference", Proceeding VLDB Endowment, 10(11), pp.1190~1201 (2017)
  14. M. Yakout, "Don’t be SCAREd : Use SCalable automatic REpairing with maximal likelihood and bounded changes", Proceedings of the 2013 ACM SIGMOD International Conference on Management of Data, 553(564), pp.553~564 (2013)
  15. M. Yakout, "Guided data repair", Proceeding VLDB Endowment, 4(5), pp.279~289 (2011)
  16. S. Krishnan, "PrivateClean : Data cleaning and differential privacy", Proceedings of the 2016 International Conference on Management of Data, pp.937~951 (2016)
  17. R. A. Popa, "CryptDB : Protecting confidentiality with encrypted query processing", in Proceedings of the twenty-third ACM symposium on operating systems principles, pp, 85(100), pp.2011~
  18. P. Mohassel, "SecureML : A system for scalable privacypreserving machine learning", IEEE Symposium on Security and Privacy (SP), pp.19~38 (2017)
  19. D. Demmler, "Aby-a framework for efficient mixed-protocol secure two-party computation", Network and Distributed System Security (NDSS), pp.59~ (2015)
  20. H. L. Dunn, "Record linkage", American Journal of Public Health Nations Health, 36(12), pp.1412~1416 (1946)
  21. I. P. Fellegi, "A Theory for Record Linkage", Journal of the American Statistical Association, 64(328), pp.1183~1210 (1969)
  22. "가명정보결합종합지원시스템"[URL]
  23. D. Vatsalan, "Handbook of Big Data Technologies", pp.1007~ (2017), Springer
  24. A. Gkoulalas-Divanis, "Modern privacy-preserving record linkage techniques : An overview", IEEE Transactions on Information Forensics and Security, 16, pp.4966~4987 (2021)[DOI]
  25. S. Gomatam, "An empirical comparison of record linkage procedures", Statistics in Medicine, 21(10), pp.1485~1496 (2002)
  26. Peter Christen, "Data matching: concepts and techniques for record linkage, entity resolution, and duplicate detection" (2012), Springer Science & Business Media
  27. A. P. Brown, "Evaluating privacy-preserving record linkage using cryptographic long-term keys and multibit trees on large medical datasets", BMC Medical Informatics and Decision Making, 17, pp.1~7 (2017)[DOI]
  28. I. Lazrig, "Privacy preserving probabilistic record linkage without trusted third party", 2018 16th Annual Conference on Privacy, Security and Trust (PST), pp.1~10 (2018)
  29. B. H. Bloom, "Space", time trade-offs in hash coding with allowable errors, 1970, pp.13(7), 422~426
  30. A novel errortolerant anonymous linking code, "R. Schnell", Social Science Research Network, ~, pp.WP-GRLC-2011-02 (2011)
  31. Christine M. O'Keefe, "Privacy-preserving data linkage protocols", Proceedings of the 2004 ACM Workshop on Privacy in the Electronic Society (WPES '04), pp.94~102 (2004), Association for Computing Machinery
  32. S. B. Dusetzina, "An Overview of Record Linkage Methods", ~ (2014)
  33. S. B. Johnson, "Using global unique identifiers to link autism collections", Journal of the American Medical Informatics Association, 17(6), pp.689~695 (2010)
  34. A. Inan, "Private record matching using differential privacy", Proceeding EDBT, pp.123~134 (2010)
  35. M. Kuzu, "Efficient privacy-aware record integration", Proceeding EDBT, pp.167~178 (2013)
  36. A. L. Potosky, "Potential for cancer related health services research using a linked Medicare-tumor registry database", Medical Care, 31(8), pp.732~748 (1993)
  37. S. J. Grannis, "Analysis of identifier performance using a deterministic linkage algorithm", Proceeding of AMIA Symposium, pp.305~309 (2002)
  38. B. McMahan, "Communication-efficient learning of deep networks from decentralized data", Artificial Intelligence and Statistics, 54 (2017)
  40. R. Xu, "Fedv : Privacy-preserving federated learning over vertically partitioned data", Proceedings of the 14th ACM Workshop on Artificial Intelligence and Security (2021)
  42. S. Stammler, "Mainzelliste SecureEpiLinker(MainSEL) : Privacypreserving record linkage using secure multi-party computation", Bioinformatics, 2020, pp.1~12 (2020)
  43. A, Southwell, "Validating a novel deterministic privacy-preserving record linkage between administrative & clinical data: applications in stroke research", International Journal of Population Data Science, 7(4), pp.1755~ (2022)[DOI]
  44. D. Morales, "Private set intersection : A systematic literature review", Computer Science Review, 49, pp.100567~ (2023)[DOI]
  46. B. McMahan, "Communication-efficient learning of deep networks from decentralized data", Artificial Intelligence and Statistics, PMLR (2017)
  47. K. Bonawitz, "Practical secure aggregation for privacy-preserving machine learning", Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security (2017)
  48. S. Truex, "A hybrid approach to privacy-preserving federated learning", Proceedings of the 12th ACM Workshop on Artificial Intelligence and Security (2019)
  49. R. Xu, "Hybridalpha : An efficient approach for privacy-preserving federated learning", Proceedings of the 12th ACM Workshop on Artificial Intelligence and Security (2019)
  50. J. Zhang, "PEFL : A privacyenhanced federated learning scheme for big data analytics", 2019 IEEE Global Communications Conference (GLOBECOM) (2019), IEEE
  51. C. Zhang, "{BatchCrypt} : Efficient homomorphic encryption for {Cross-Silo} federated learning", 2020 USENIX Annual Technical Conference (USENIX ATC 20) (2020)
  52. G. Xu, "Verifynet : Secure and verifiable federated learning", IEEE Transactions on Information Forensics and Security, 15, pp.911~926 (2019)
  53. X. Guo, "VeriFL : Communication-Efficient and Fast Verifiable Aggregation for Federated Learning", IEEE Transactions on Information Forensics and Security, 16, pp.1736~1751 (2020)
  54. H. Fereidooni, "SAFELearn : Secure aggregation for private federated learning", 2021 IEEE Security and Privacy Workshops (SPW) (2021), IEEE
  55. J. Park, "Privacy-preserving federated learning using homomorphic encryption", Applied Sciences, 12(2), pp.734~ (2022)
  56. Y. A. Shin, "Securing a local training dataset size in federated learning", IEEE Access, 10, pp.104135~104143 (2022)
  57. J. Ma, "Privacy-preserving federated learning based on multi-key homomorphic encryption", International Journal of Intelligent Systems, 37(9), pp.5880~5901 (2022)
  58. Y. Cheng, "Federated learning for privacy-preserving AI", Communications of the ACM, 63(12), pp.33~36 (2020)
  59. M. G. Poirot, "Split Learning for collaborative deep learning in healthcare", NeurIPS (2019)
  60. "Google Research" (2017)[URL]
  62. A. Gascón, "Secure linear regression on vertically partitioned datasets", International Association for Cryptologic Research Cryptology ePrint Archive, 892 (2016)
  64. B. Gu, "Federated doubly stochastic kernel learning for vertically partitioned data", Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (2020)
  67. K. Cheng, "Secureboost : A lossless federated learning framework", IEEE Intelligent Systems, 36(6), pp.87~98 (2021)
  68. Q. Zhang, "Secure bilevel asynchronous vertical federated learning with backward updating", Proceedings of the AAAI Conference on Artificial Intelligence, 35(12) (2021)
  69. S. Warnat-Herresthal, "Swarm Learning for decentralized and confidential clinical machine learning", Nature, 594, pp.265~270 (2021)
  70. O. L. Saldanha, "Swarm learning for decentralized artificial intelligence in cancer histopathology", Nature Medicine, 28(6), pp.1232~1239 (2022)
  71. H. Basak, "A union of deep learning and swarm-based optimization for 3D human action recognition", Scientific Reports, 12(1), pp.5494~ (2022)
  72. F. Wang, "A reinforcement learning level-based particle swarm optimization algorithm for large-scale optimization", Information Sciences, 602, pp.298~312 (2022)
  73. M. Al-Rubaie, "Privacy-preserving machine learning : Threats and solutions", IEEE Security &Privacy, 17(2), pp.49~58 (2019)
  75. G. A. Kaissis, "Secure, privacy-preserving and federated machine learning in medical imaging", Nature Machine Intelligence, 2(6), pp.305~311 (2020)

관련 논문AI추천

연합 학습 환경에서의 Task-Specific Adaptive Differential Privacy 메커니즘 평가 방안 연구최윤호, 우타리예바 아쎔한국정보보호학회 · 정보보호학회논문지2024 · 제34권 제1호연합학습의 국방 AI 적용 방안박찬호, 박태현, 서중하, 이영준국방보안연구소 · 국방과 보안2023 · 제5권 제2호연합학습 환경에서의 적대적 훈련 클라이언트 비율에 따른 전역 모델 강건성 비교김현일, 류권상, 서창호, 정수용, 조규찬한국정보보호학회 · 정보보호학회논문지2025 · 제35권 제4호인공지능 서비스의 프라이버시 침해에 관한 연구이한성, 조영복한국실천공학교육학회 · 실천공학교육논문지2025 · 제17권 제5호보안 환경에서 병렬학습된 딥러닝 모형의 통합에 관한 연구이종락한국컴퓨터정보학회 · 한국컴퓨터정보학회논문지2021 · 제26권 제12호하이브리드 클라우드 환경 기반 연합학습 시스템 아키텍처 연구김광기, 김진술, 신혜주, 오상원, 오승민한국지식정보기술학회 · 한국지식정보기술학회 논문지2023 · 제18권 제2호

이 간행물의 다른 논문 (8건)

공간정보 표준기반 스마트시티 프레임워크정국식, 구경철, 고은비pp. 1-12합 기반의 전건부를 가지는 뉴로-퍼지 시스템 설계한창욱, 이돈규pp. 13-17도메인 정보 모델 간의 상호운용성 확보를 위한 모델 레지스트리 활용방안홍상기, 최원욱pp. 18-25디지털 트윈의 모듈화 기법 설계: 면 제조 공정을 중심으로권찬우, 송석현pp. 26-33다중 컴파트먼트 비밀공유 기법유민수, 최철훈pp. 34-40
원문 보기