arXiv Machine Learning

Occupancy-based Quantile Risk Control

Occupancy-based Quantile Risk Control (OQRC) is a new method that extends conformal risk control to quantile-based risk measures. It partitions the loss space using ordered calibration losses, estimates the distribution of test losses in each bin, and upper-bounds the risk by the maximum loss per bin. The approach guarantees finite-sample validity, achieving tight risk control bounds that converge at a rate of σ(n^{-1/2}) and reducing the risk gap by up to 78.64% in experiments.

Hugging Face Trending Papers

Sep 2

Occupancy-based Quantile Risk Control

Occupancy-based Quantile Risk Control (OQRC) is a new method that extends conformal risk control to quantile-based risk measures while avoiding excessive conservatism and providing rigorous finite-sample guarantees. It works by partitioning the loss space using ordered calibration losses, estimating the distribution of test losses in each bin, and bounding the risk by the maximum loss in each bin. The authors prove that OQRC achieves tight risk control bounds with a finite-sample guarantee that converges at a rate of π(n−½), and experiments show it can reduce the risk gap by up to 78.64% on common benchmarks.

benchmarks

arXiv Machine Learning

Sep 14

Quantile-based Loss Filtering for Outlier-Robust Stochastic Gradient Descent

arXiv:2609. 13040v1 Announce Type: new Abstract: We study loss-based filtering for finite-sum optimization with a subset of corrupted component functions whose gradients may be highly unreliable.

By Jamie Haddock, Anna Ma, Elizaveta Rebrova

arXiv Machine Learning

Jun 11

Calibrating Decision Robustness via Inverse Conformal Risk Control

arXiv:2510. 07750v3 Announce Type: replace-cross Abstract: Robust optimization safeguards decisions against uncertainty by optimizing against worst-case scenarios, yet their effectiveness hinges on a prespecified robustness level that is often chosen ad hoc, leading to either insufficient protection or overly conservative and costly solutions.

By Wenbin Zhou, Shixiang Zhu

arXiv Machine Learning

Sep 22

Conformalized Quantile Regression and Minimax Limits of Fixed-Score Calibration under Known Covariate Shift

arXiv:2609.24929v1 Announce Type: cross Abstract: In this paper, we study nonasymptotic $L^p$ error bounds for interval length and conditional coverage in split conformalized quantile regression (CQR...

By Rustam Isaev, Anton Conrad, Denis Belomestny, Eric Moulines, Sergey Samsonov

benchmarks

arXiv Machine Learning

Sep 2

Online simultaneous inference for quantiles via smoothed stochastic gradient descent

arXiv:2505.13299v2 Announce Type: replace-cross Abstract: This paper considers the estimation of quantiles via a smoothed version of the stochastic gradient descent (SGD) algorithm. By smoothing the...

By Likai Chen, Georg Keilbar, Wei Biao Wu

efficiency

arXiv Machine Learning

Sep 1

A Unified Perspective on Conformal Prediction and Wasserstein Distributionally Robust Optimization for Uncertainty Quantification

arXiv:2608.29789v1 Announce Type: cross Abstract: Uncertainty quantification from finite data is central to machine learning, optimization, and automation systems, where decisions must remain reliabl...

By Kehan Long, Yiqi Zhao, Pol Mestres, Lars Lindemann, Nikolay Atanasov, Jorge Cort\'es

arXiv Machine Learning

Jul 30

Simultaneous Coverage and Efficiency Guarantee in Online Conformal Prediction

arXiv:2607. 26577v1 Announce Type: new Abstract: Adaptive conformal inference (ACI) of Gibbs and Cand{\`e}s and its variants are the standard approach to online conformal prediction under distribution shift, but they suffer from three fundamental limitations.

By Rahul Vaze

benchmarks safety

Hugging Face Trending Papers

Jun 21

Distribution-Aware Robust Bilevel Optimization: Quantile-Guided Huber Updates in Two-Timescale Stochastic Approximation

Bilevel optimization (BLO) is fundamental to hierarchical decision-making but suffers from critical instability under heavy-tailed stochastic noise. Existing variance-reduction techniques typically rely on myopic magnitude checks, which fail to distinguish informative geometric signals from impulsive outliers.

reinforcement-learning benchmarks

arXiv Machine Learning

Aug 18

Non-Crossing Deep Quantile Regression for Distributional Survival Prediction

arXiv:2608. 16864v1 Announce Type: cross Abstract: In survival analysis the way covariates act on the risk of an event often differs between early and late failure times, yet hazard- and mean-based summaries collapse this variation into a single number.

By Shuai Huang, Zhe Qu, Zhaowei Hua, Guohao Shen, Rui Tang, Hongtu Zhu

llms

arXiv Machine Learning

Sep 11

Certifying Lower Bounds for Risk-Sensitive Reinforcement Learning under Adversarial State Perturbations

arXiv:2609. 10866v1 Announce Type: new Abstract: Reinforcement learning (RL) agents deployed in real-world environments are often vulnerable to adversarial perturbations in state observations, creating risks in safety-critical applications.

By Tong Li, Saunak Kumar Panda, Yisha Xiang

agents reinforcement-learning safety

arXiv AI

Jun 4

Scenario Generation for Risk-Aware Reinforcement Learning with Probably Approximately Safe Guarantees

arXiv:2606. 04812v1 Announce Type: cross Abstract: Guaranteeing safety is critical to the deployment of reinforcement learning (RL) agents in the real-world, especially as policies learned using deep RL may demonstrate susceptibility to transition perturbations that result in unknown or unsafe behaviour.

By Mohit Prashant, Arvind Easwaran

agents reinforcement-learning safety

arXiv Machine Learning

Sep 10

Understanding Uncertainty Sampling via Equivalent Loss

arXiv:2307.02719v5 Announce Type: replace Abstract: Uncertainty sampling is a classical active-learning strategy, yet the statistical objective induced by its query rule is often implicit. We introdu...

By Shang Liu, Xiaocheng Li

Related stories