AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Practical Adversarial Multivalid Conformal Prediction

Osbert Bastani, Varun Gupta, Christopher Jung, Georgy Noarov, Ramya Ramalingam, Aaron Roth

NeurIPS 2022 · Learning theory and algorithms › Prediction and decision losses · approach: Conformal prediction sets · 7 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use MVP rather than RAPSImageNet classification with ResNet-152 · ImageNet validationmarginal coverage: 0.90523 -> 0.902133No clear effect
Use MVP rather than RAPSImageNet classification with ResNet-152 · ImageNet validationaverage prediction-set size: 2.0506 -> 2.13986Hurt
Use MVP instead of split conformal while training the regression model on all past observationsSynthetic linear regression · Synthetic linear regression, marginal coverageempirical coverageImproved
Use MVP instead of split conformal while training the regression model on all past observationsSynthetic linear regression · Synthetic linear regression, marginal coverageaverage interval widthImproved
Use MVP rather than naive split conformal for intersecting-group calibrationSynthetic regression with 20 intersecting groups · Synthetic regression, group-wise coveragegroup-wise coverageImproved
Use MVP rather than conservative groupwise conformalSynthetic regression with 20 intersecting groups · Synthetic regression, group-wise coverageaverage interval widthImproved
Use MVP rather than split conformal under unknown state-to-state shiftFolktables income prediction · California to Pennsylvania covariate shiftempirical coverageImproved
Use MVP rather than ACI for intersecting subgroupsGARCH stock-volatility prediction · Noisy stock-return groupsgroup-wise coverageImproved
Use threshold-calibrated MVP rather than marginal-only ACI on monotonically increasing scoresStock-return adversarial ordering · Scores increasing from 0.0 to 0.5average interval width: 1.839 -> 0.526Improved
Use normalized rather than un-normalized surrogate score function (ablation)MVP · Experiments across reported settingsperformanceNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy