Adaptive Deep Metric Learning with harmonized loss and nearest proxy alignment for low-dimensional representation

Computer Science

• Approximation (Algorithm) 100%
• Embedding Space 50%
• Information Retrieval 50%
• Learning Framework 50%
• Adaptive Parameter 50%
• Training Process 50%
• Dimensionality Reduction 50%
• Local Structure 50%
• Subclasses 50%