Distribution Privacy Under Function Recoverability

نویسندگان

چکیده

A user generates $n$ independent and identically distributed data random variables with a probability mass function that must be guarded from querier. The querier recover, prescribed accuracy, given of the each query responses upon eliciting them user. chooses devises to maximize distribution privacy as gauged by (Kullback-Leibler) divergence between former querier’s best estimate it based on responses. Considering an arbitrary function, basic achievable lower bound for is provided does not depend corresponds worst-case privacy. Worst-case equals logsum cardinalities inverse atoms under number summands decreasing recovers improving accuracy. Next, upper (converse) (achievability) bounds privacy, dependent , are developed. improves latter so suitable assumptions; both converge grows. converse achievability proofs identify explicit strategies

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Recoverability of Joint Distribution from Missing Data

A probabilistic query may not be estimable from observed data corrupted by missing values if the data are not missing at random (MAR). It is therefore of theoretical interest and practical importance to determine in principle whether a probabilistic query is estimable from missing data or not when the data are not MAR. We present an algorithm that systematically determines whether the joint pro...

متن کامل

Discrete Distribution Estimation under Local Privacy

The collection and analysis of user data drives improvements in the app and web ecosystems, but comes with risks to privacy. This paper examines discrete distribution estimation under local privacy, a setting wherein service providers can learn the distribution of a categorical statistic of interest without collecting the underlying data. We present new mechanisms, including hashed k-ary Random...

متن کامل

Discrete Distribution Estimation under Local Privacy

As argued in the proof sketch of Theorem 2, it suffices to show that r ,ε,k,n (Q) obeys the data processing inequality. is the minimax risk in the non-private setting.

متن کامل

Discrete Distribution Estimation under Local Privacy

The collection and analysis of user data drives improvements in the app and web ecosystems, but comes with risks to privacy. This paper examines discrete distribution estimation under local privacy, a setting wherein service providers can learn the distribution of a categorical statistic of interest without collecting the underlying data. We present new mechanisms, including hashed k-ary Random...

متن کامل

The Bernstein Mechanism: Function Release under Differential Privacy

We investigate the problem of privately answering queries on databases consisting of points in [0, 1]. We prove the following results. First, we show that there exists a computationally efficient ε-differentially private mechanism that releases a query class parametrized by additively separable Hölder continuous functions. Second, we show that, if the query class is instead parametrized by addi...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: IEEE Transactions on Information Theory

سال: 2022

ISSN: ['0018-9448', '1557-9654']

DOI: https://doi.org/10.1109/tit.2022.3140317