AI Paper Atlas

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

Measures of distortion for machine learning

Leena Chennuru Vankadara, Ulrike von Luxburg

NeurIPS 2018 · Learning theory and algorithms › Prediction and decision losses · approach: Data attribution and recourse · 8 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 sigma-distortion instead of average distortion as the measureIsomap embeddings of Gaussian data · Embedding dimension vs distortiondistortion decreases with embedding dimensionImproved
use sigma-distortion or epsilon-distortion instead of worst-case, average, normalized average, k-localIsomap embeddings of gamma and beta data · Original dimension 10 to 100, fixed embedding dimensiondistortion increases with original dimensionImproved
use sigma-distortion instead of other measures under additive noiseMixture-of-Gaussians data with noise in R20 · Variance of noise vs distortiondistortion increases with noise varianceImproved
increase variance of additive noiseSVM and kNN on Isomap embeddings · Noisy Gaussian mixture classificationclassification lossHurt
use sigma-distortion or epsilon-distortion as indicator instead of worst-case, average, k-localEmbeddings from PCA, GPLVM, Isomap, MVU, SPE · Distortion vs classification loss (kNN and SVM)monotone relation with classification lossImproved
use other embedding algorithms (PCA, MVU) instead of Isomapnoise experiments · Noise vs classification lossclassification loss trendNo clear effect
use sigma-distortion instead of other measures regarding variability across repetitionsIsomap embeddings · Original dimension vs distortion error barsstd over 10 repetitionsImproved

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