arXiv:2303.02770
공식 코드 · Papers with Code
Universal distribution of the empirical coverage in split conformal prediction
원문 초록
When split conformal prediction operates in batch mode with exchangeable data, we determine the exact distribution of the empirical coverage of prediction sets produced for a finite batch of future observables, as well as the exact distribution of its almost sure limit when the batch size goes to infinity. Both distributions are universal, being determined solely by the nominal miscoverage level and the calibration sample size, thereby establishing a criterion for choosing the minimum required calibration sample size in applications.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
Enhancing Adversarial Robustness with Conformal Prediction: A Framework for Guaranteed Model Reliability
Nearness of Neighbors Attention for Regression in Supervised Finetuning
Efficient Robust Conformal Prediction via Lipschitz-Bounded Networks
Conformal Prediction for Zero-Shot Models
Optimal Conformal Prediction under Epistemic Uncertainty
Foundation models for time series forecasting: Application in conformal prediction
토론
이 논문과 연결된 포럼 글이 아직 없습니다.