cs.LGJun 26, 2026

Halt Fast! Early Stopping for Certified Robustness

Authors: Andrew C. CullenPaul MontagueBenjamin I. P. Rubinstein

Abstract

Randomized Smoothing (RS) provides rigorous robustness guarantees for neural networks without architectural constraints, yet its adoption is limited by extreme computational costs. Standard RS requires tens of thousands of model evaluations per input and forces practitioners to commit to fixed sample sizes a priori. In this work, we present a novel meta-learning framework for anytime-valid certified robustness that adaptively deploys computational resources. By using a lightweight meta-learner to predict image-specific priors for a sequential E-process, we achieve a 20-fold reduction in sample complexity compared to traditional methods while maintaining rigorous statistical guarantees. Beyond raw efficiency, we demonstrate how anytime-validity enables adaptively allocating compute based upon application-specific risk thresholds, a form of resource triage impossible under classic certification frameworks. That this is achievable while also providing similar certification performance demonstrates that our approach provides a pathway for real-time, safety-critical certification deployments.

Explore similar work

CardsList
  1. Laplace-Bridged Randomized Smoothing for Fast Certified Robustness

    Apr 27, 2026Miao Lin, MD Saifur Rahman Mazumder, Feng Yu +2SmoothingNoise-Aware

  2. RRISE: Robust Radius Inference via a Surrogate Estimator

    Jun 1, 2026Jong-Ik Park, Shreyas Chaudhari, Carlee Joe-Wong +1SmoothingSurrogate Performance