Communication-Efficient Parallelization Strategy for Deep Convolutional Neural Network Training

Computer Science

• Gradient Descent 100%
• Deep Convolutional Neural Networks 100%
• Parallelization 100%
• Neural Network Training 100%
• Neural Network Model 66%
• Residual Neural Network 66%
• VGG-16 66%
• Supercomputer 33%
• Time Complexity 33%
• Convolutional Neural Network 33%
• Interprocess Communication 33%
• Design Strategy 33%
• Complexity Analysis 33%
• Convergence Rate 33%