cs.LGNov 4, 2026

Meta-Frieren: Beyond Single Level Online Learning

Authors: Vahan Arsenyan, Arnak Dalalyan

Organizations: Department of Statistics CREST, ENSAE, IP Paris Rte de Saclay, Palaiseau 91120, France

Abstract

We study a two-stage online optimization problem: at each round, a meta-learner first selects an agent-environment pair, then the chosen agent interacts with the environment, and receives a reward. We introduce Meta-Frieren, a general framework for maximizing cumulative reward in this setting. Our analysis yields a simple UCB-style criterion for meta-level selection and recovers prior guarantees as special cases while extending them to broader settings. We instantiate Meta-Frieren with GP-UCB under Bayesian and frequentist assumptions. We also derive, to our knowledge, the first high-probability regret bound for Feel-Good Thompson Sampling, specialize it to finite and parametric function classes, and integrate it into Meta-Frieren. Experiments support our theoretical results and demonstrate that Meta-Frieren can be successfully applied to hyperparameter optimization across multiple model classes.

Figures & tables

Explore similar work

CardsList
  1. ICICLE: In-Context Inference for Curve-Learning Extrapolation

    Nov 18, 2026Laurits Fredsgaard, Michael Riis Andersen, Mikkel N. SchmidtHyperparameter OptimizationLearning Curve Prediction

  2. Online Learning with Gradient-Variation Interval Regret

    Jun 2, 2026Yan-Feng Xie, Shuche Wang, Peng Zhao +1