Active Privacy-utility Trade-off Against a Hypothesis Testing Adversary
This addresses privacy-utility trade-offs for users sharing data in online services, but it is incremental as it applies existing RL methods to a specific modeling scenario.
The paper tackles the problem of a user sequentially releasing data to balance privacy and utility, where personal information includes a secret variable to keep private and a useful variable to disclose. They formulate this as a Markov decision process and solve it numerically using advantage actor-critic deep reinforcement learning, achieving trade-offs between correct detection probability and mutual information.
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).