An Empirical Study of Self-Supervised Learning with Wasserstein Distance

Keyphrases

• Self-supervised Learning 100%
• Wasserstein Distance 100%
• Probability Model 33%
• Softmax 16%
• Distance Model 16%
• Cosine Similarity 16%
• Tree Structure 8%
• Regularization Method 8%
• Model Performance 8%
• Two-tree 8%
• Total Variation 8%
• Training Model 8%
• Simplicial 8%
• Training Procedure 8%
• Representation Learning 8%
• CIFAR-100 8%
• CIFAR-10 8%
• Embedding Vector 8%
• L1 Distance 8%
• Stable Training 8%
• Similarity-based Representation 8%
• Self-supervised Learning Methods 8%
• SimSiam 8%
• SimCLR 8%

Mathematics

• Softmax Function 16%