Certified Decisions

2025-02-25 · Isaiah Andrews, Jiafeng Chen

Hypothesis tests and confidence intervals are ubiquitous in empirical research, yet their connection to subsequent decision-making is often unclear. We develop a theory of certified decisions that pairs recommended decisions with inferential guarantees. Specifically, we attach P-certificates -- upper bounds on loss that hold with probability at least $1-\alpha$ -- to recommended actions. We show that such certificates allow "safe," risk-controlling adoption decisions for ambiguity-averse downstream decision-makers. We further prove that it is without loss to limit attention to P-certificates arising as minimax decisions over confidence sets, or what Manski (2021) terms "as-if decisions with a set estimate." A parallel argument applies to E-certified decisions obtained from e-values in settings with unbounded loss.

📄 PDF Abstract BibTeX arXiv:2502.17830

Code (0)

등록된 구현이 없습니다.

Tasks

Decision Making

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Attention 설명 없음

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

UC-Search: Risk-Aware Test-Time Search for Delayed Constrained Time-Series Control

2026-06-24 · Xibai Wang arxiv

Time-series deployments often need delayed feasible decisions, not only accurate forecasts. UC-Search is a trace-only retained-search layer for delayed constrained control: a frozen backbone emits forecasts or action sco…

Certified Circuits: Stability Guarantees for Mechanistic Circuits

2026-02-26 · Alaa Anani, Tobias Lorenz, Bernt Schiele, Mario Fritz 외 arxiv

Understanding how neural networks arrive at their predictions is essential for debugging, auditing, and deployment. Mechanistic interpretability pursues this goal by identifying circuits--minimal subnetworks responsible …

CertVLA: Certified Defense against Physical Visual Attacks for Vision-Language-Action Models

2026-08-21 · Hui Lu, Zhijie Peng, Yuqi Lin, Zaijia Yang 외 arxiv

Vision-Language-Action (VLA) policies are vulnerable to localized physical perturbations, yet existing certified patch defenses target discrete labels and cannot directly certify continuous, temporally correlated actions…

LipCache: A Local Inference Proxy with Certified Caching for Edge Image Classification Service

2026-08-13 · Zhengzhe Xiang, Yinlin Chen, Fuli Ying, Binbin Zhou 외 arxiv

As edge-side vision services continue to expand toward low-latency, high-throughput scenarios, reducing the inference cost of vision models without sacrificing reliability has become a central concern. Existing semantic …

Image Classification

Certified Adversarial Robustness for Deep Reinforcement Learning

2019-10-28 · Björn Lütjens, Michael Everett, Jonathan P. How

Deep Neural Network-based systems are now the state-of-the-art in many robotics tasks, but their application in safety-critical domains remains dangerous without formal guarantees on network robustness. Small perturbatio…

Adversarial Robustness Collision Avoidance Deep Reinforcement Learning reinforcement-learning+2