SV-SAE: Layer-Wise Pruning for Autoencoder Based on Link Contributions

Computer Science

• Autoencoder 100%
• Sparse Autoencoder 100%
• Autoencoder Layer 100%
• Sparsity 60%
• Unsupervised Learning 20%
• Deep Neural Network 20%
• Dimensionality Reduction 20%
• Essential Feature 20%
• Experimental Result 20%
• Performance Degradation 20%
• Multilayer Perceptron 20%
• Individual Contribution 20%
• Feature Extraction 20%
• cooperative game theory 20%

Engineering

• Autoencoder 100%
• Sparsity 30%
• Constrainedness 20%
• Experimental Result 10%
• Input Data 10%
• Performance Degradation 10%
• Feature Extraction 10%
• Dimensionality 10%
• Essential Feature 10%
• Deep Neural Network 10%
• Perceptron 10%