BROWSE
Detail
Conference Paper
A Method on Neural Network Optimization Deployment Frameworks for Lightweight Target Devices
Cited 1 time in 
Share 
Share
- Authors
- Jaebok Park, Kyunghee Lee, Jiyoung Kwak, Changsik Cho
- Issue Date
- 2023-10
- Citation
- International Conference on Information and Communication Technology Convergence (ICTC) 2023, pp.1286-1288
- Publisher
- IEEE
- Language
- English
- Type
- Conference Paper
- Abstract
- The development of artificial intelligence neural networks requires specialized knowledge in the industry. The process of creating and deploying neural networks is very difficult for software developers who lack knowledge. Therefore, there is a need for tools that can easily develop neural network applications in industries. In order to meet these needs, this paper proposes a deployment optimized for a target device by automatically generating a specification-based neural network. First, the user simply selects the desired system and neural network requirements. The proposed framework utilizes user specifications to generate the desired neural network. Next, template code is generated based on the generated neural network. Finally, the neural network is deployed to target devices. In this way, we present a method to build an artificial intelligence neural network application deployment easily and quickly.
- KSP Keywords
- Network requirements, Neural Network Applications, Neural network optimization, Template code, application deployment, artificial intelligence, need for, neural network(NN), software developers, specification-based