Accelerating Convolutional Neural Network Inference in Split Computing: An In-Network Computing Approach

Keyphrases

• In-network Computing 100%
• Convolutional Neural Network 100%
• Deep Neural Network 100%
• Split Computing 100%
• Neural Network Inference 100%
• Mobile Devices 75%
• Computing Node 75%
• Convolutional Neural Network Model 50%
• Placement Scheme 50%
• Inference Latency 50%
• Edge Server 50%
• Optimization Problem 25%
• Promising Technology 25%
• Memory Resource 25%
• Arithmetic Operations 25%
• Low Latency 25%
• Evaluation Results 25%
• Resource Constraints 25%
• Simple Arithmetic 25%
• Multiple Layers 25%
• Computing Approach 25%
• Device Server 25%
• Network Device 25%
• Packet Processing 25%
• Limited Computing Resources 25%
• Max Pooling Layer 25%
• Edge Switch 25%
• Device Edge 25%
• Complex Deep Neural Networks 25%
• Computational Tasks 25%
• Programmable Switches 25%
• Mobile Edge 25%

Computer Science

• Neural Network Model 100%
• Convolutional Neural Network 100%
• Network Inference 100%
• Deep Neural Network 66%
• Mobile Device 50%
• Edge Server 33%
• Optimization Problem 16%
• Memory Resource 16%
• Arithmetic Operation 16%
• Complex Process 16%
• Evaluation Result 16%
• Resource Constraint 16%
• Computing Resource 16%
• Packet Processing 16%
• Computational Task 16%