Accelerating Deep Learning Inference via Model Parallelism and Partial Computation Offloading

Mathematics

• Parallelism 100%
• Learning 79%
• Internet of Things 59%
• Partial 58%
• Neural Networks 51%
• Strategy 38%
• Model 28%
• Neural Network Model 27%
• Scheduling 22%
• Flexibility 20%
• Task Assignment 19%
• Particle Swarm Optimization 15%
• Path 14%
• Complex Networks 12%
• Accelerate 12%
• Time Complexity 11%
• Trade-offs 11%
• Update 10%
• Optimal Solution 9%
• Computing 6%
• Performance 6%

Engineering & Materials Science

• Deep learning 78%
• Deep neural networks 74%
• Internet of things 27%
• Scheduling 15%
• Edge computing 13%
• Network layers 12%
• Particle swarm optimization (PSO) 9%