AI Paper Atlas

Research map / Label-efficient and robust learning / Test-time adaptation

Cold Case: The Lost MNIST Digits

Chhavi Yadav, Leon Bottou

NeurIPS 2019 · Label-efficient and robust learning › Test-time adaptation · approach: Extrapolation and single-sample adaptation · 63 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
Evaluate classifiers on the reconstructed lost test examples rather than the official MNIST test setKNN, SVM, MLP, and CNN classifiers · QMNIST50K versus MNIST testmisclassification rate: 2.95% -> 3.19%; 1.47% -> 1.8%; 1.61% -> 2.02%; 0.37% -> 0.53%Hurt
Train LeNet-5 on reconstructed QMNIST instead of MNISTLeNet-5 · MNIST testmisclassification rate: $0.82\%$ -> $0.81\%No clear effect
Train LeNet-5 on reconstructed QMNIST instead of MNISTLeNet-5 · QMNIST10Kmisclassification rate: $0.81\%$ -> $0.80\%No clear effect
Train KNN on QMNIST instead of MNISTKNN (k=3) · QMNIST50Kmisclassification rate: $3.19\%$ -> $3.19\%No clear effect
Train SVM on QMNIST instead of MNISTRBF SVM (C=10, g=0.02) · QMNIST50Kmisclassification rate: $1.8\%$ -> $1.8\%No clear effect
Train MLP on QMNIST instead of MNISTMLP (800 hidden units) · QMNIST50Kmisclassification rate: $2.02\%$ -> $2\%No clear effect
Select a model using MNIST test performanceClassifiers evaluated on MNIST and QMNIST50K · MNIST and QMNIST50Kclassifier rankingImproved
Use paired comparisons on the same test examples instead of comparing absolute error-rate confidence intervalsClassifier comparisons · MNIST testmisclassification comparison sensitivity: four additional misclassifications out of 10,000Improved
Choose 1100 rather than 700 hidden units for the MLPMLP · QMNIST50Kmisclassification rateMixed

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