AI Paper Atlas

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

Synbols: Probing Learning Algorithms with Synthetic Datasets

Alexandre Lacoste, Pau Rodríguez López, Frederic Branchaud-Charron, Parmida Atighehchian, Massimo Caccia, Issam Hadj Laradji, Alexandre Drouin, Matthew Craddock, Laurent Charlin, David Vázquez

NeurIPS 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Contrastive representation theory · 5 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
add data augmentation (ResNet-12 -> ResNet-12+) (ablation)ResNet-12 · Synbols Less Variations 100kaccuracy: 39.41 -> 57.42Improved
add data augmentation (WRN-28-4 -> WRN-28-4+) (ablation)WRN-28-4 · Synbols Less Variations 100kaccuracy: 23.10 -> 68.42Improved
replace flattening with global average pooling (ablation)Conv-4 · Synbols Korean 100kaccuracy: 1.62 -> 33.58Improved
add camouflage to the background/foregroundWRN-28-4+ · Synbols Default 100kaccuracy: 97.41 -> 95.55Hurt
increase training set from 100k to 1MMLP · Synbols Defaultaccuracy: 14.83 -> 63.05Improved
stratified rotation split instead of i.i.d.WRN-28-4+ · Synbols Defaultaccuracy: 97.41 -> 91.81Hurt
stratified rotation split instead of i.i.d.Conv-4 Flat · Synbols Defaultaccuracy: 68.51 -> 5.54Hurt
stratified font split instead of i.i.d.ResNet-12 · Synbols Defaultaccuracy: 95.43 -> 94.62No clear effect
stratified x-translation split instead of i.i.d.Conv-4 Flat · Synbols Defaultaccuracy: 68.51 -> 44.78Hurt
stratified x-translation split instead of i.i.d.ResNet-12+ · Synbols Defaultaccuracy: 97.16 -> 96.84No clear effect
test on unseen characters for font predictionWRN-28-4+ · Synbols Less Variationsaccuracy: 68.42 -> 35.41Hurt
use P-Entropy instead of BALD query functionactive learning VGG-16 · 10% Missingnegative log likelihood: 1.24 -> 2.02Hurt

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