JMSNAS: Joint Model Split and Neural Architecture Search for Learning over Mobile Edge Networks
This work addresses efficient DNN deployment for mobile edge computing, offering a practical solution for resource-constrained environments, though it appears incremental as it builds on existing split machine learning techniques.
The paper tackles the challenge of deploying deep neural networks on mobile edge networks by proposing JMSNAS, a framework that jointly optimizes model splitting and neural architecture search to match network constraints, achieving superior performance over state-of-the-art methods in terms of accuracy and latency trade-offs.
The main challenge to deploy deep neural network (DNN) over a mobile edge network is how to split the DNN model so as to match the network architecture as well as all the nodes' computation and communication capacity. This essentially involves two highly coupled procedures: model generating and model splitting. In this paper, a joint model split and neural architecture search (JMSNAS) framework is proposed to automatically generate and deploy a DNN model over a mobile edge network. Considering both the computing and communication resource constraints, a computational graph search problem is formulated to find the multi-split points of the DNN model, and then the model is trained to meet some accuracy requirements. Moreover, the trade-off between model accuracy and completion latency is achieved through the proper design of the objective function. The experiment results confirm the superiority of the proposed framework over the state-of-the-art split machine learning design methods.