The quantization algorithm compresses the original network by reducing the numerical bit width of the model,which improves the computation speed. Because different layers have different redundancy and sensitivity to d...The quantization algorithm compresses the original network by reducing the numerical bit width of the model,which improves the computation speed. Because different layers have different redundancy and sensitivity to databit width. Reducing the data bit width will result in a loss of accuracy. Therefore, it is difficult to determinethe optimal bit width for different parts of the network with guaranteed accuracy. Mixed precision quantizationcan effectively reduce the amount of computation while keeping the model accuracy basically unchanged. In thispaper, a hardware-aware mixed precision quantization strategy optimal assignment algorithm adapted to low bitwidth is proposed, and reinforcement learning is used to automatically predict the mixed precision that meets theconstraints of hardware resources. In the state-space design, the standard deviation of weights is used to measurethe distribution difference of data, the execution speed feedback of simulated neural network accelerator inferenceis used as the environment to limit the action space of the agent, and the accuracy of the quantization model afterretraining is used as the reward function to guide the agent to carry out deep reinforcement learning training. Theexperimental results show that the proposed method obtains a suitable model layer-by-layer quantization strategyunder the condition that the computational resources are satisfied, and themodel accuracy is effectively improved.The proposed method has strong intelligence and certain universality and has strong application potential in thefield of mixed precision quantization and embedded neural network model deployment.展开更多
Improving capacity and reducing delay are the most challenging topics in wireless ad hoc networks. Nodes that equip multiple radios working on different channels simultaneously permit ef-fective utility of frequency s...Improving capacity and reducing delay are the most challenging topics in wireless ad hoc networks. Nodes that equip multiple radios working on different channels simultaneously permit ef-fective utility of frequency spectrum and can also reduce interference. In this paper, after analyzing several current protocols in Multi-Radio Multi-Channel (MR-MC) ad hoc networks, a new multi-channel routing metric called Integrative Route Metric (IRM) is designed. It takes channel load, inter-flow, and intra-flow interference into consideration. In addition, an MR-MC routing protocol based on Interference-Aware and Channel-Load (MR-IACL) is also presented. The MR-IACL can assign channels and routings for nodes according to channel load and interference degree of links, and optimize channel distribution dynamically to satisfy the features of topology changing and traffic frequent fluctuation during network running. The simulation results show that the new protocol outperforms others in terms of network throughput, end-to-end delay, routing overhead, and network lifetime.展开更多
文摘The quantization algorithm compresses the original network by reducing the numerical bit width of the model,which improves the computation speed. Because different layers have different redundancy and sensitivity to databit width. Reducing the data bit width will result in a loss of accuracy. Therefore, it is difficult to determinethe optimal bit width for different parts of the network with guaranteed accuracy. Mixed precision quantizationcan effectively reduce the amount of computation while keeping the model accuracy basically unchanged. In thispaper, a hardware-aware mixed precision quantization strategy optimal assignment algorithm adapted to low bitwidth is proposed, and reinforcement learning is used to automatically predict the mixed precision that meets theconstraints of hardware resources. In the state-space design, the standard deviation of weights is used to measurethe distribution difference of data, the execution speed feedback of simulated neural network accelerator inferenceis used as the environment to limit the action space of the agent, and the accuracy of the quantization model afterretraining is used as the reward function to guide the agent to carry out deep reinforcement learning training. Theexperimental results show that the proposed method obtains a suitable model layer-by-layer quantization strategyunder the condition that the computational resources are satisfied, and themodel accuracy is effectively improved.The proposed method has strong intelligence and certain universality and has strong application potential in thefield of mixed precision quantization and embedded neural network model deployment.
基金Supported by the National Natural Science Foundation of China (No. 60873195, No. 61070220)the Research Fund for the Doctoral Program of Higher Education of China (No. 20090111110002)
文摘Improving capacity and reducing delay are the most challenging topics in wireless ad hoc networks. Nodes that equip multiple radios working on different channels simultaneously permit ef-fective utility of frequency spectrum and can also reduce interference. In this paper, after analyzing several current protocols in Multi-Radio Multi-Channel (MR-MC) ad hoc networks, a new multi-channel routing metric called Integrative Route Metric (IRM) is designed. It takes channel load, inter-flow, and intra-flow interference into consideration. In addition, an MR-MC routing protocol based on Interference-Aware and Channel-Load (MR-IACL) is also presented. The MR-IACL can assign channels and routings for nodes according to channel load and interference degree of links, and optimize channel distribution dynamically to satisfy the features of topology changing and traffic frequent fluctuation during network running. The simulation results show that the new protocol outperforms others in terms of network throughput, end-to-end delay, routing overhead, and network lifetime.
