Privacy-Aware Time-Series Data Sharing With Deep Reinforcement Learning

Erdemir, Ecenaz; Dragotti, Pier Luigi; Gündüz, Deniz

doi:10.1109/tifs.2020.3013200

Cited by 39 publications

(27 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Accordingly, we consider MI as a utility measure; that is, the user wants to maximize the MI between the useful hypothesis and the observations by the time the adversary reaches the prescribed confidence level on the secret. MI is commonly used both as a privacy and a utility measure in the literature [4,8,15] The MI between U and (Z T , A T ) over time T is given by…”

Section: Pomdp Formulationmentioning

confidence: 99%

“…Privacy is an important concern for the adoption of many IoT services, and there is a growing demand from consumers to keep their personal information private. Privacy has been widely studied in the literature [1][2][3][4][5][6][7][8][9][10], and a vast number of privacy measures have been introduced, including differential privacy [1], mutual information (MI) [2][3][4][5][6][7][8], total variation distance [11], maximal leakage [12,13], and guessing leakage [14], to count a few.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Active Privacy-Utility Trade-Off Against A Hypothesis Testing Adversary

Erdemir

Dragotti

Gündüz

2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

We consider a user releasing her data containing some personal information in return of a service. We model user's personal information as two correlated random variables, one of them, called the secret variable, is to be kept private, while the other, called the useful variable, is to be disclosed for utility. We consider active sequential data release, where at each time step the user chooses from among a finite set of release mechanisms, each revealing some information about the user's personal information, i.e., the true hypotheses, albeit with different statistics. The user manages data release in an online fashion such that maximum amount of information is revealed about the latent useful variable, while the confidence for the sensitive variable is kept below a predefined level. For the utility, we consider both the probability of correct detection of the useful variable and the mutual information (MI) between the useful variable and released data. We formulate both problems as a Markov decision process (MDP), and numerically solve them by advantage actor-critic (A2C) deep reinforcement learning (RL).

show abstract

Section: Pomdp Formulationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Active Privacy-Utility Trade-Off Against A Hypothesis Testing Adversary

Erdemir

Dragotti

Gündüz

2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

show abstract

“…The privacy notions studied therein include k-anonymity [16], [17], (extended) differential privacy [18]- [20], and distortion privacy [21], [22], which all differ from the information-theoretic privacy measure studied in this paper. The works of [23], [24] recently studied the information-theoretic privacy measure in location-privacy protection mechanisms, and their privacy metric was defined by the mutual information between the released data and the true traces. In this paper's language, it can be viewed as the case when privacy is always ON.…”

Section: B Related Workmentioning

confidence: 99%

ON-OFF Privacy Against Correlation Over Time

Naim

Rouayheb

2021

IEEE Trans.Inform.Forensic Secur.

View full text Add to dashboard Cite

We consider the problem of ON-OFF privacy in which a user is interested in the latest message generated by one of n sources available at a server. The user has the choice to turn privacy ON or OFF depending on whether he wants to hide his interest at the time or not. The challenge of allowing the privacy to be toggled between ON and OFF is that the user's online behavior is correlated over time. Therefore, the user cannot simply ignore the privacy requirement when privacy is OFF.We represent the user's correlated requests by an n-state Markov chain. Our goal is to design ON-OFF privacy schemes with optimal download rate that ensure privacy for past and future requests. We devise a polynomial-time algorithm to construct an ON-OFF privacy scheme. Moreover, we present an upper bound on the achievable rate. We show that the proposed scheme is optimal and the upper bound is tight for some special families of Markov chains. We also give an implicit characterization of the optimal achievable rate as a linear programming (LP).Index Terms-Information-theoretic privacy, private information retrieval, Markov chains I. INTRODUCTION A. MotivationIn the current data-driven world, users' information is always being collected online, and its privacy has become a significant concern. Many users wish to keep private their personal information, such as their age, sex, political views, health disorders, etc. Significant research has been devoted to study algorithms that preserve users' privacy. Some of the proposed approaches include applying anonymization techniques [1], differential privacy algorithms [2], and private information retrieval methods [3].Privacy, however, comes at a cost. Privacy-preserving algorithms typically incur higher overheads in terms of computation, memory, and delay. These incurred costs motivate one to think of privacy as an expensive commodity and, therefore, to allow the user to request it, i.e., turn privacy ON, only when needed; otherwise, turn it OFF. The user may choose to switch between privacy being ON and OFF depending on several criteria, such as location (country, workplace vs. home, etc.), network connection (public or private network), devices (shared vs. personal machines) being used, or service quality (privacy-preserving algorithms typically induce more overheads), to name a few.At a conceptual level, ON-OFF privacy algorithms enable privacy to be switched between ON and OFF whenever desired. One of the main challenges in designing such algorithms

show abstract

“…Therefore, research studies focus on multiple domains separately. For example, privacy and IoT [36][37][38], privacy in AI, and machine learning [39,40] privacy in Process Mining [41][42][43][44]. This targeted attention lagged in paying concerted attention to the overall context of the issue.…”

Section: Ethical Issuesmentioning

confidence: 99%

Multilevel Privacy Assurance Evaluation of Healthcare Metadata

2021

View full text Add to dashboard Cite

Healthcare providers are legally bound to ensure the privacy preservation of healthcare metadata. Usually, privacy concerning research focuses on providing technical and inter-/intra-organizational solutions in a fragmented manner. In this wake, an overarching evaluation of the fundamental (technical, organizational, and third-party) privacy-preserving measures in healthcare metadata handling is missing. Thus, this research work provides a multilevel privacy assurance evaluation of privacy-preserving measures of the Dutch healthcare metadata landscape. The normative and empirical evaluation comprises the content analysis and process mining discovery and conformance checking techniques using real-world healthcare datasets. For clarity, we illustrate our evaluation findings using conceptual modeling frameworks, namely e3-value modeling and REA ontology. The conceptual modeling frameworks highlight the financial aspect of metadata share with a clear description of vital stakeholders, their mutual interactions, and respective exchange of information resources. The frameworks are further verified using experts’ opinions. Based on our empirical and normative evaluations, we provide the multilevel privacy assurance evaluation with a level of privacy increase and decrease. Furthermore, we verify that the privacy utility trade-off is crucial in shaping privacy increase/decrease because data utility in healthcare is vital for efficient, effective healthcare services and the financial facilitation of healthcare enterprises.

show abstract

Privacy-Aware Time-Series Data Sharing With Deep Reinforcement Learning

Cited by 39 publications

References 38 publications

Active Privacy-Utility Trade-Off Against A Hypothesis Testing Adversary

Active Privacy-Utility Trade-Off Against A Hypothesis Testing Adversary

ON-OFF Privacy Against Correlation Over Time

Multilevel Privacy Assurance Evaluation of Healthcare Metadata

Contact Info

Product

Resources

About