Besov function approximation and binary classification on low-dimensional manifolds using convolutional residual networks

Keyphrases

• Binary Classification 100%
• Low-dimensional Manifolds 100%
• Function Approximation 100%
• Data Dimensions 100%
• Low-dimensional Structures 100%
• Convolutional Residual Network 100%
• Sample Complexity 100%
• Besov Functions 100%
• High-dimensional Data 50%
• Deep Learning 50%
• Deep Neural Network 50%
• Network Architecture 50%
• Intrinsic Dimension 50%
• Statistical Theory 50%
• Excess Risk 50%
• Arbitrary Accuracy 50%
• Smoothness Parameter 50%
• Logistic Risk 50%

Mathematics

• Binary Classification 100%
• Approximation Function 100%
• Dimensional Manifold 100%
• Dimensional Structure 100%
• Residual Network 100%
• Dimensional Data 50%
• Classification Problem 50%
• Approximates 50%
• Manifold 50%
• Deep Neural Network 50%
• Deep Learning Method 50%
• Approximation of Function 50%
• Statistical Theory 50%
• Excess Risk 50%
• Arbitrary Accuracy 50%