Near-Optimal Model Discrimination with Non-Disclosure
Let θ_0,θ_1 ∈ℝ^d be the population risk minimizers associated to some loss ℓ: ℝ^d ×𝒵→ℝ and two distributions ℙ_0,ℙ_1 on 𝒵. We pose the following question: Given i.i.d. samples from ℙ_0 and ℙ_1, what sample sizes are sufficient and necessary to distinguish between the two hypotheses θ^* = θ_0 and θ^* = θ_1 for given θ^* ∈{θ_0, θ_1}? Making the first steps towards answering this question in full generality, we first consider the case of a well-specified linear model with squared loss. Here we provide matching upper and lower bounds on the sample complexity, showing it to be min{1/Δ^2, √(r)/Δ} up to a constant factor, where Δ is a measure of separation between ℙ_0 and ℙ_1, and r is the rank of the design covariance matrix. This bound is dimension-independent, and rank-independent for large enough separation. We then extend this result in two directions: (i) for the general parametric setup in asymptotic regime; (ii) for generalized linear models in the small-sample regime n ≤ r and under weak moment assumptions. In both cases, we derive sample complexity bounds of a similar form, even under misspecification. Our testing procedures only access θ^* through a certain functional of empirical risk. In addition, the number of observations that allows to reach statistical confidence in our tests does not allow to "resolve" the two models – that is, recover θ_0,θ_1 up to O(Δ) prediction accuracy. These two properties allow to apply our framework in applied tasks where one would like to identify a prediction model, which can be proprietary, while guaranteeing that the model cannot be actually inferred by the identifying agent.
READ FULL TEXT