AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Discriminative Unsupervised Feature Learning with Exemplar Convolutional Neural Networks

Alexey Dosovitskiy, Philipp Fischer, Jost Tobias Springenberg, Martin Riedmiller, Thomas Brox

TPAMI 2016 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 805 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
increase surrogate classes up to about 8000 (ablation)Exemplar-CNN 64c5-64c5-128f · STL-10, CIFAR-10, Caltech-101classification accuracy (%): 50 -> 8000 classes helps; beyond 8000 no change or decreaseMixed
add more samples per surrogate class (1 to 300) (ablation)Exemplar-CNN 64c5-64c5-128f · STL-10classification accuracy (%): saturates at around 100 samplesImproved
add translation, color and contrast transformations to seed augmentation (ablation)Exemplar-CNN 64c5-64c5-128f · STL-10, CIFAR-10, Caltech-101classification accuracy (%)Improved
add rotation and scaling transformations (ablation)Exemplar-CNN 64c5-64c5-128f · STL-10, CIFAR-10, Caltech-101classification accuracy (%)No clear effect
add occlusion, affine transformation, Gaussian noise to augmentation (ablation)Exemplar-CNN · classificationclassification accuracy (%)No clear effect
add clustering to build surrogate classes (ablation)Exemplar-CNN 92c5-256c5-512c5-1024f · STL-10classification accuracy (%): 74.2 -> 75.4Improved
train seed patches from the test dataset itself instead of STL-10 (Caltech-101 seeds, Caltech-101 test) (ablation)Exemplar-CNN 64c5-64c5-128f · Caltech-101classification accuracy (%): 79.8 -> 80.0No clear effect
train on CIFAR-10 seed patches instead of STL-10 (ablation)Exemplar-CNN 64c5-64c5-128f · STL-10classification accuracy (%): 67.1 -> 64.5Hurt
add a fourth convolutional layer (ablation)Exemplar-CNN 64c5-64c5-64c5-128f · STL-10classification accuracy (%): 68.5 -> 64.7Hurt
add a third convolutional layer (ablation)Exemplar-CNN 64c5-64c5-128f · STL-10classification accuracy (%): 67.1 -> 68.5Improved
widen layers from 64-64-128 to 128-256-512 (ablation)Exemplar-CNN 64c5-64c5-128f · STL-10classification accuracy (%): 67.1 -> 71.2Improved
use 7x7 instead of 5x5 first-layer filters (ablation)Exemplar-CNN 64c5-64c5-128f · STL-10classification accuracy (%): 67.1 -> 66.3No 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