High-Probability Minimax Adaptive Estimation in Besov Spaces via Online-to-Batch
Preprint
Abstract
We study nonparametric regression over Besov spaces from noisy observations under sub-exponential noise, aiming to achieve minimax-optimal guarantees on the integrated squared error that hold with high probability and adapt to the unknown noise level. To this end, we propose a wavelet-based online learning algorithm that dynamically adjusts to the observed gradient noise by adaptively clipping it at an appropriate level, eliminating the need to tune parameters such as the noise variance or gradient bounds. As a by-product of our analysis, we derive high-probability adaptive regret bounds that scale with the $\ell_1$-norm of the competitor.Finally, in the batch statistical setting, we obtain adaptive and minimax-optimal estimation rates for Besov spaces via a refined online-to-batch conversion. This approach carefully exploits the structure of the squared loss in combination with self-normalized concentration inequalities.Keywords
All Related Versions
This publication has 8 references indexed in Scilit:
- Stochastic online convex optimization. Application to probabilistic time series forecastingElectronic Journal of Statistics, 2024
- Empirical entropy, minimax regret and minimax riskBernoulli, 2017
- Introduction to Online Convex OptimizationFoundations and Trends® in Optimization, 2016
- Online Learning and Online Convex OptimizationFoundations and Trends in Machine Learning, 2012
- Exponential inequalities for self-normalized martingales with applicationsThe Annals of Applied Probability, 2008
- On the Generalization Ability of On-Line Learning AlgorithmsIEEE Transactions on Information Theory, 2004
- Wavelets, Approximation, and Statistical Applications1998
- On Minimax Wavelet EstimatorsApplied and Computational Harmonic Analysis, 1996