AI Paper Atlas

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

Multi-label Contrastive Predictive Coding

Jiaming Song, Stefano Ermon

NeurIPS 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Partial and multi-label learning · 9 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 ML-CPC with J_0.01 instead of CRDCIFAR-100 distillation, WRN-40-2 to WRN-16-2 · CIFAR-100top-1 test accuracy (%): 75.48 -> 75.83Improved
Use ML-CPC with J_0.01 instead of CRDCIFAR-100 distillation, resnet110 to resnet20 · CIFAR-100top-1 test accuracy (%): 71.46 -> 71.27Hurt
Use ML-CPC with J_0.01 instead of CRDCIFAR-100 distillation, resnet32x4 to resnet8x4 · CIFAR-100top-1 test accuracy (%): 75.51 -> 75.37Hurt
Use smaller alpha (J_0.05) rather than alpha 1.0 ML-CPC (ablation)CIFAR-100 distillation, WRN-40-2 to WRN-40-1 · CIFAR-100top-1 test accuracy (%): 74.18 -> 74.27Improved
Use smaller alpha (J_0.05) rather than alpha 1.0 ML-CPC (ablation)CIFAR-100 distillation, ResNet50 to MobileNetV2 · CIFAR-100top-1 test accuracy (%): 68.80 -> 70.04Improved
Replace CPC with ML-CPC using a high-to-low alpha curriculumUnsupervised representation learning · CIFAR-10, 200 epochstop-1 accuracy: 83.28 -> 86.16Improved
Use the J_10.0 to J_0.1 curriculum rather than fixed alpha 1.0 ML-CPC (ablation)Unsupervised representation learning · CIFAR-100, 200 epochstop-1 accuracy: 61.80 -> 64.05Improved
Use the J_10.0 to J_0.1 curriculum rather than fixed alpha 1.0 ML-CPC (ablation)Unsupervised representation learning · CIFAR-100, 1000 epochstop-1 accuracy: 70.85 -> 70.03Hurt
Replace CPC with ML-CPC using a high-to-low alpha curriculumUnsupervised representation learning · ImageNet, 30 epochsTop1: 43.45 -> 43.86Improved
Replace CPC with ML-CPC using a high-to-low alpha curriculumUnsupervised representation learning · ImageNet, 30 epochsTop5: 67.42 -> 67.67Improved
Use ML-CPC instead of CPCMutual information estimation with 128 positive and negative samples · Correlated Gaussians, 20 dimensionscomputation time for 200 updates (seconds): 1.15 ± 0.06 -> 1.14 ± 0.04No clear effect
Reduce alpha for ML-CPC (ablation)Mutual information estimation · Correlated Gaussians, 20 dimensionsestimation biasImproved

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