Optimal Sampling Density for Nonparametric Regression
This provides a robust and interpretable active learning strategy for regression tasks, addressing the challenge of efficient sampling in nonparametric settings.
The paper tackles the problem of active learning for regression by deriving the optimal training density that minimizes generalization error for local polynomial smoothing, resulting in a closed-form solution that outperforms existing model-agnostic methods in simulations.
We propose a novel active learning strategy for regression, which is model-agnostic, robust against model mismatch, and interpretable. Assuming that a small number of initial samples are available, we derive the optimal training density that minimizes the generalization error of local polynomial smoothing (LPS) with its kernel bandwidth tuned locally: We adopt the mean integrated squared error (MISE) as a generalization criterion, and use the asymptotic behavior of the MISE as well as the locally optimal bandwidths (LOB) - the bandwidth function that minimizes MISE in the asymptotic limit. The asymptotic expression of our objective then reveals the dependence of the MISE on the training density, enabling analytic minimization. As a result,we obtain the optimal training density in a closed-form. The almost model-free nature of our approach thus helps to encode the essential properties of the target problem, providing a robust and model-agnostic active learning strategy. Furthermore, the obtained training density factorizes the influence of local function complexity, noise level and test density in a transparent and interpretable way. We validate our theory in numerical simulations, and show that the proposed active learning method outperforms the existing state-of-the-art model-agnostic approaches.