Papers
Topics
Authors
Recent
Search
2000 character limit reached

PEaRL: Personalized Privacy of Human-Centric Systems using Early-Exit Reinforcement Learning

Published 9 Mar 2024 in cs.LG, cs.CR, and cs.HC | (2403.05864v2)

Abstract: In the evolving landscape of human-centric systems, personalized privacy solutions are becoming increasingly crucial due to the dynamic nature of human interactions. Traditional static privacy models often fail to meet the diverse and changing privacy needs of users. This paper introduces PEaRL, a system designed to enhance privacy preservation by tailoring its approach to individual behavioral patterns and preferences. While incorporating reinforcement learning (RL) for its adaptability, PEaRL primarily focuses on employing an early-exit strategy that dynamically balances privacy protection and system utility. This approach addresses the challenges posed by the variability and evolution of human behavior, which static privacy models struggle to handle effectively. We evaluate PEaRL in two distinct contexts: Smart Home environments and Virtual Reality (VR) Smart Classrooms. The empirical results demonstrate PEaRL's capability to provide a personalized tradeoff between user privacy and application utility, adapting effectively to individual user preferences. On average, across both systems, PEaRL enhances privacy protection by 31%, with a corresponding utility reduction of 24%.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (58)
  1. Armand Ahadi-Sarkani and Salma Elmalaki. 2021. Adas-rl: Adaptive vector scaling reinforcement learning for human-in-the-loop lane departure warning. In Proceedings of the First International Workshop on Cyber-Physical-Human System Design and Implementation. 13–18.
  2. Thermal environmental conditions for human occupancy. Inc.Atlanta, GA, USA (2010).
  3. The sleep-evoked decrease of body temperature. Sleep 16, 2 (1993), 93–99.
  4. Lauren Bass. 2019. The Concealed Cost of Convenience: Protecting Personal Data Privacy in the Age of Alexa. Fordham Intell. Prop. Media & Ent. LJ 30 (2019), 261.
  5. Repel: A utility-preserving privacy system for iot-based energy meters. In 2020 IEEE/ACM Fifth International Conference on Internet-of-Things Design and Implementation (IoTDI). IEEE, 79–91.
  6. Kelly Caine. 2009. Exploring everyday privacy behaviors and misclosures. Georgia Institute of Technology.
  7. João Miguel Reis Araújo Proença Cambeiro. 2019. The Human in the loop in Cyber-Physical Systems: the case of Building Automation. Ph. D. Dissertation.
  8. campbellsci. 2022. black globe thermometer. Retrieved June 10, 2022 from https://www.campbellsci.com/blackglobe
  9. Robert G Carroll. 2006. Elsevier’s Integrated Physiology E-Book. Elsevier Health Sciences.
  10. Deep reinforcement learning for Internet of Things: A comprehensive survey. IEEE Communications Surveys & Tutorials 23, 3 (2021), 1659–1692.
  11. Salma Elmalaki. 2021. Fair-iot: Fairness-aware human-in-the-loop reinforcement learning for harnessing human variability in personalized iot. In Proceedings of the International Conference on Internet-of-Things Design and Implementation. 119–132.
  12. Internet of Personalized and Autonomous Things (IoPAT): Smart Homes Case Study. In Proceedings of the 1st ACM International Workshop on Smart Cities and Fog Computing. ACM, 35–40.
  13. Sentio: Driver-in-the-loop forward collision warning using multisample reinforcement learning. In Proceedings of the 16th ACM Conference on Embedded Networked Sensor Systems. 28–40.
  14. EMOTIV. 2020. EMOTIV EPOC+ 14 Channel Mobile Brainwear. https://www.emotiv.com/product/emotiv-epoc-14-channel-mobile-eeg/. Last accessed June 2023.
  15. Privacy-Aware Time-Series Data Sharing With Deep Reinforcement Learning. IEEE Transactions on Information Forensics and Security 16 (2020), 389–401.
  16. Paul J Eslinger and Antonio R Damasio. 1985. Severe disturbance of higher cognition after bilateral frontal lobe ablation: patient EVR. Neurology 35, 12 (1985), 1731–1731.
  17. Poul O Fanger. 1970. Thermal comfort. Analysis and applications in environmental engineering. Thermal comfort. Analysis and applications in environmental engineering. (1970).
  18. Aviezri S Fraenkel. 2004. Complexity, appeal and challenges of combinatorial games. Theoretical Computer Science 313, 3 (2004), 393–415.
  19. Privacy amplification via shuffling for linear contextual bandits. In International Conference on Algorithmic Learning Theory. PMLR, 381–407.
  20. Local differential privacy for regret minimization in reinforcement learning. Advances in Neural Information Processing Systems 34 (2021), 10561–10573.
  21. Michael Gerber. 2014. energyplus energy Simulation Software. (2014).
  22. Cooperative inverse reinforcement learning. Advances in neural information processing systems 29 (2016).
  23. Safe exploration for reinforcement learning.. In ESANN. 143–148.
  24. Jakub P Hlávka. 2020. Security, privacy, and information-sharing aspects of healthcare artificial intelligence. In Artificial Intelligence in Healthcare. Elsevier, 235–270.
  25. Dynamic privacy budget allocation improves data efficiency of differentially private gradient descent. In Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency. 11–35.
  26. Experimenting with electromagnetism using augmented reality: Impact on flow student experience and educational effectiveness. Computers & Education 71 (2014), 1–13.
  27. T Io. 2011. economic and Business Dimensions online advertising, Behavioral targeting, and Privacy. CommunICatIons of the aCm 54, 5 (2011).
  28. On the tradeoff between privacy and utility in collaborative intrusion detection systems-a game theoretical approach. In Proceedings of the Hot Topics in Science of Security: Symposium and Bootcamp. 45–51.
  29. Wooyoung Jung and Farrokh Jazizadeh. 2017. Towards integration of doppler radar sensors into personalized thermoregulation-based control of HVAC. In Proceedings of the 4th ACM International Conference on Systems for Energy-Efficient Built Environments. ACM, 21.
  30. KhanAcademy. 2021a. AP College Biology. https://www.khanacademy.org/science/ap-biology. Accessed: 2021-10-07.
  31. KhanAcademy. 2021b. AP College Physics. https://www.khanacademy.org/science/ap-physics-2. Accessed: 2021-10-07.
  32. KhanAcademy. 2021c. Organic Chemistry. https://www.khanacademy.org/science/organic-chemistry. Accessed: 2021-10-07.
  33. RAPID-RL: A Reconfigurable Architecture with Preemptive-Exits for Efficient Deep-Reinforcement Learning. In 2022 International Conference on Robotics and Automation (ICRA). IEEE, 7492–7498.
  34. Human thermoregulation and measurement of body temperature in exercise and clinical settings. Annals Academy of Medicine Singapore 37, 4 (2008), 347.
  35. Privacy-preserving bandits. Proceedings of Machine Learning and Systems 2 (2020), 350–362.
  36. MATLAB. 2022. Thermal Model of a House. Retrieved June 10, 2022 from https://www.mathworks.com/help/simulink/slref/thermal-model-of-a-house.html
  37. Split computing and early exiting for deep learning applications: Survey and research challenges. Comput. Surveys 55, 5 (2022), 1–30.
  38. W. J. McKeachie and M. Svinicki. 2006. McKeachie’s teaching tips: Strategies, research, and theory for college and university teachers (12th ed.). Houghton-Mifflin.
  39. Secure Data Communication Using Information Hiding and Encryption Algorithms. In 2022 Second International Conference on Artificial Intelligence and Smart Energy (ICAIS). IEEE, 1448–1452.
  40. A lightweight and inexpensive in-ear sensing system for automatic whole-night sleep stage monitoring. In Proceedings of the 14th ACM Conference on Embedded Network Sensor Systems. 230–244.
  41. Stress-induced hyperthermia and anxiety: pharmacological validation. European journal of pharmacology 463, 1-3 (2003), 117–132.
  42. Liberty Adams Omonkhoa. 2021. A review of the issues and challenges in IoT security using machine learning techniques. (2021).
  43. How You Act Tells a Lot: Privacy-Leaking Attack on Deep Reinforcement Learning. AAMAS (2019).
  44. To Share or Not to Share: On Location Privacy in IoT Sensor Data. In 2022 IEEE/ACM Seventh International Conference on Internet-of-Things Design and Implementation (IoTDI). IEEE, 128–140.
  45. Model explanations with differential privacy. In Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency. 1895–1904.
  46. Maria Petrescu and Anjala S Krishen. 2018. Analyzing the analytics: data privacy concerns. , 41–43 pages.
  47. Active preference-based learning of reward functions.
  48. Sensors-based wearable systems for monitoring of human movement and falls. IEEE Sensors Journal 12, 3 (2012), 658–670.
  49. Decision making support for privacy data upload in smart home. In Adjunct Proceedings of the 2019 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2019 ACM International Symposium on Wearable Computers. 214–217.
  50. Erudite: Human-in-the-loop iot for an adaptive personalized learning system. IEEE Internet of Things Journal (2023).
  51. adaparl: Adaptive privacy-aware reinforcement learning for sequential-decision making human-in-the-loop systems. arXiv preprint arXiv:2303.04257 (2023).
  52. Branchynet: Fast inference via early exiting from deep neural networks. In 2016 23rd international conference on pattern recognition (ICPR). IEEE, 2464–2469.
  53. Detecting Learner Drowsiness Based on Facial Expressions and Head Movements in Online Courses. In Proceedings of the 25th International Conference on Intelligent User Interfaces Companion. 124–125.
  54. The New York Times. 2022. The Best Smart Thermostat. https://www.nytimes.com/wirecutter/reviews/the-best-thermostat/. Accessed: 2022-06-10.
  55. Baoxiang Wang and Nidhi Hegde. 2019. Privacy-preserving q-learning with functional noise in continuous spaces. Advances in Neural Information Processing Systems 32 (2019).
  56. Enhancing Privacy in Federated Learning via Early Exit. In Proceedings of the 5th workshop on Advanced tools, programming languages, and PLatforms for Implementing and Evaluating algorithms for Distributed systems. 1–5.
  57. Attribute privacy: Framework and mechanisms. In Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency. 757–766.
  58. FAIRO: Fairness-aware Adaptation in Sequential-Decision Making for Human-in-the-Loop Systems. arXiv preprint arXiv:2307.05857 (2023).
Citations (1)
View on Semantic Scholar

Summary

No one has generated a summary of this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Sign Up to Summarize

Paper to Video (Beta)

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Sign Up to Generate

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Generate Now

Continue Learning

We haven't generated follow-up questions for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Generate Now

Collections

Sign up for free to add this paper to one or more collections.

User Circle Single Streamline Icon: https://streamlinehq.com Sign Up

Tweets

Sign up for free to view the 1 tweet with 0 likes about this paper.

User Circle Single Streamline Icon: https://streamlinehq.com Sign Up for Free