Deep compressive offloading: Speeding up neural network inference by trading edge computation for network latency

Keyphrases

• Neural Network 100%
• Latency 100%
• Network Latency 100%
• Edge Computation 100%
• Accuracy Loss 100%
• Edge Computing 100%
• Offloading System 100%
• Neural Network Inference 100%
• Sensing Applications 50%
• Lower End 50%
• Recent Advances 50%
• Compressed Sensing Theory 50%
• Computing Resources 50%
• Speech Recognition 50%
• Computational Demand 50%
• Computer Vision 50%
• Perfect Reconstruction 50%
• System Application 50%
• Network Bandwidth 50%
• Tiny Size 50%
• Comprehensive Evaluation 50%
• Data Transmission Rate 50%
• Low Latency 50%
• Vision Recognition 50%
• End Device 50%
• Edge Devices 50%
• Deep Learning 50%
• Intelligent Services 50%
• Background Traffic 50%
• IoT Devices 50%
• Edge Server 50%
• Network Offload 50%
• Intelligent IoT 50%
• Computation Offloading 50%
• IoT Sensing 50%
• Theory Learning 50%

Computer Science

• Neural Network 100%
• Network Latency 100%
• Network Inference 100%
• Edge Computing 50%
• Internet-Of-Things 50%
• Compressive Sensing 25%
• Speech Recognition 25%
• Perfect Reconstruction 25%
• Speed-up 25%
• Building-Blocks 25%
• Computing Resource 25%
• Network Bandwidth 25%
• Transferring Data 25%
• Edge Server 25%
• Computer Vision 25%
• Deep Learning Method 25%