Dynamic Hierarchical Neural Network Offloading in IoT Edge Networks

Wassim Seifeddine (1) , Cédric Adjih (1) , Nadjib Achir (2)

1 TRiBE - inTeRnet BEyond the usual
2 L2TI - Laboratoire de Traitement et Transport de l'Information

In recent developments in machine learning, a trend has emerged where larger models achieve better performance. At the same time, deploying these models in real-life scenarios is difficult due to the parallel trend of pushing them on end-users or IoT devices with strong resource limitations. In this work, we develop a novel technique for executing parts of a single model successively through multiple devices (IoT, edge, cloud) while respecting each device's resource limitations. For that, we introduce a new offloading mechanism where, during computation, a decision can be made to offload work, together with the ability to exit early in the computation with intermediate results. The decision itself is tuned through Deep Q-Learning.

Mots clés: Performance evaluation; Wireless communication; Q-learning; Computational modeling; Neural networks; Machine learning; Market research

Domaines: Réseaux et télécommunications [cs.NI]

Wassim Seifeddine, Cédric Adjih, Nadjib Achir. Dynamic Hierarchical Neural Network Offloading in IoT Edge Networks. PEMWN 2021 - 10th IFIP International Conference on Performance Evaluation and Modeling in Wireless and Wired Networks, Nov 2021, Ottawa / Virtual, Canada. pp.1-6, ⟨10.23919/PEMWN53042.2021.9664700⟩. ⟨hal-03533536⟩