AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

Morpho-MNIST: Quantitative Assessment and Diagnostics for Representation Learning

Daniel C. Castro, Jeremy Tan, Bernhard Kainz, Ender Konukoglu, Ben Glocker

JMLR 2019 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Disentangled and discrete latent models · 35 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 on locally perturbed digits (swelling, fractures) instead of plain test digitskNN · MNIST digit recognitionaccuracy (%): 96.25 -> 95.22Hurt
evaluate on locally perturbed digits instead of plain test digitsSVM · MNIST digit recognitionaccuracy (%): 95.71 -> 92.47Hurt
evaluate on locally perturbed digits instead of plain test digitsMLP · MNIST digit recognitionaccuracy (%): 97.97 -> 93.15Hurt
evaluate on locally perturbed digits instead of plain test digitsCNN (LeNet-5) · MNIST digit recognitionaccuracy (%): 98.95 -> 95.33Hurt
train InfoGAN on data with random thinning and thickening plus an extra continuous code (InfoGAN-B vs InfoGAN-A)InfoGAN · Overall MIG on inferred codesmutual information gap: 0.0582 -> 0.1748Improved
add two binary latent codes and train on swollen and fractured digits (InfoGAN-C vs InfoGAN-A)InfoGAN · Overall MIG on inferred codesmutual information gap: 0.0582 -> 0.0507Hurt
reduce GAN latent dimension from 64 to 2GAN · Morphometric MMD test vs test dataMMD^2_l (x10^-3): 1.458 -> 8.876Hurt
use a CNN instead of kNN for abnormality detectionabnormality detection · Morpho-MNIST localaccuracy (%): 65.10 -> 97.53Improved
use a CNN instead of kNN for thickness regressionthickness regression · Morpho-MNIST globalRMSE: 0.467 -> 0.279Improved

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