AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

Novel Uncertainty Quantification Through Perturbation-Assisted Sample Synthesis

Yifei Liu, Rex Shen, Xiaotong Shen

TPAMI 2024 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Bayesian decision-making under uncertainty

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
increase the inference sample size n from 2,050 to 10,000PASS-DDPM null FID · CIFAR-10FID: 49.75 -> 32.58Improved
use DCGAN instead of DDPM as candidate generatorPAI image test · CIFAR-10 (n = 10,000)FID: 32.57 -> 76.17Hurt
use GLOW instead of DDPM as candidate generatorPAI image test · CIFAR-10 (n = 10,000)FID: 32.57 -> 61.01Hurt
add perturbation of size tau up to 1.0 to PASS (ablation)PASS-TabDDPM · Prediction interval simulationFID: 0.024 -> 0.024No clear effect
add perturbation of size tau up to 1.0 to PASS (ablation)PASS-TabDDPM · Prediction interval simulation2-Wasserstein: 1.298 -> 1.298No clear effect
use PAI intervals instead of conformal inference intervalsPAI prediction intervals · Simulation (alpha = 0.05)interval length: about 68% of intervals shorterImproved
use PAI instead of conformal inference for coveragePAI prediction intervals · Simulation (alpha = 0.05)coverage probability: median 0.95, average 0.9Mixed
use PASS null distribution instead of asymptotic Gaussian for sentiment testPAI sentiment test · IMDBKS test P-value: P-value .004, .002, .000Improved
compare identical prompts versus different promptsPAI coherence test · Stable Diffusion promptsFID score: .544 -> 14.250Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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