Finite-Sample Binary Hypothesis Testing via R\'enyi Divergences: Strong Converse and Local Privacy

Roberto Bruno; Adrien Vandenbroucque; Amedeo Roberto Esposito

In the authors' words

We study asymmetric simple binary hypothesis testing between $H_{0} : P_{0}^{n}$ and $H_{1} : P_{1}^{n}$ , based on $n$ independent and identically distributed observations. Leveraging a variational representation of R\'enyi divergence of order $α$ , we derive our main result: a finite-sample converse with $α > 1$ . The bound uses both directions of the divergence $D_{α} (P_{1} ∥ P_{0})$ and $D_{α} (P_{0} ∥ P_{1})$ , tensorises under product measures, and contains familiar data-processing converses as boundary cases. For comparison, we apply the same variational approach to general $f$ -divergences and specialise it to total variation, $E_{γ}$ , Hellinger, and Kullback Leibler divergences, thereby recovering familiar converses within a unified framework. Together with an achievability bound involving R\'enyi divergence with $α \in (0, 1)$ , the main converse recovers the phase transition of the optimal Type II error under the exponentially decaying Type I error constraint $ε_{n} = e^{- n r}$ . Under regularity conditions, the optimal Type II error vanishes exponentially when $r < D (P_{1} ∥ P_{0})$ and converges exponentially fast to one when $r > D (P_{1} ∥ P_{0})$ . We also derive sample-complexity bounds and extend both the converse and achievability analyses to locally differentially private observations, quantifying the cost of privacy and recovering the non-private achievability bound as the privacy constraint vanishes.

Main resultThe abstract does not state a limitation.

Finite-Sample Binary Hypothesis Testing via R\'enyi Divergences: Strong Converse and Local Privacy

In the authors' words

More on Statistics that day

Bayesian inference, on-line forecasting and model choice for large VAR models with Cholesky stochastic volatility

Short-term rental market occupancy - daily time series for 2017-2022 on 500 markets worldwide

The Spectra of the Henze-Zirkler and Henze-Wagner Operators for BHEP Tests

Pooling Sequential Evidence Across Hypotheses: Rate-Optimal Multiple Testing at a Fixed Horizon

Rolling Conformal Prediction in Sequential Model Training