Performance optimization on model synchronization in parallel stochastic gradient descent based SVM

Computer Science

• Training Model 100%
• Support Vector Machine 100%
• Performance Optimization 100%
• Gradient Descent 100%
• Support Vector Machine 66%
• Distributed Model 66%
• Large Data Set 33%
• Training Process 33%
• Speed-up 33%
• Fast Convergence 33%
• Parallel Process 33%
• Sequential Algorithm 33%

Keyphrases

• Performance Optimization 100%
• Support Vector Machine 100%
• Model Synchronization 100%
• Parallel Stochastic Gradient Descent 100%
• Training Model 33%
• Communication Time 33%
• Synchronization Frequency 33%
• Support Vector Machine Algorithm 22%
• Distributed Model 22%
• Training Process 11%
• Fast Convergence 11%
• Stochastic Gradient Descent 11%
• Model Optimization 11%
• Sequential Algorithm 11%
• Global Synchronization 11%
• Distributed Support Vector Machines 11%
• Testing Accuracy 11%
• Mini-batch Size 11%
• High Frequency Model 11%

Engineering

• Support Vector Machine 100%
• Gradient Descent 100%
• Processed Data 33%
• Batch Size 33%