Reduced-memory training and deployment of deep residual networks by stochastic binary quantization

Computer Science

• Deep Residual Network 100%
• binarization 100%
• Wide Residual Networks 66%
• Energy Efficient 33%
• Deep Neural Network 33%
• Hardware Implementation 33%
• Image Classification 33%
• Floating Point 33%
• Backpropagation Algorithm 33%
• Single Precision 33%
• Point Implementation 33%
• Trained Network 33%