HARNESSING THE POWER OF INFINITELY WIDE DEEP NETS ON SMALL-DATA TASKS

Mathematics

• Neural Network 100%
• Variance 100%
• Running Time 100%
• Kernel Method 100%
• Training Sample 100%
• Transfer Learning 100%
• Gold Standard 100%

Computer Science

• Support Vector Machine 100%
• Testbed 66%
• Random Decision Forest 33%
• Efficient Algorithm 33%
• Gradient Descent 33%
• Image Classification 33%
• Kernel Method 33%
• Classification Task 33%
• Regression Task 33%
• Gold Standard 33%
• Learning Rate 33%
• Transfer Learning 33%
• Training Sample 33%
• Residual Neural Network 33%
• Trained Neural Network 33%

Keyphrases

• Linear SVM 12%
• Squeeze-and-excitation Network (SENet) 12%