Computational Offloading with Deep Supervised Learning for Edge enabled IoT
Abstract
Edge enabled IoT tries to offer better QoE to the IoT users. But, the rapid growth of user demands leads to the lower level of QoE. In order to improve the level of QoE, we use a deep deterministic learning-based approach; namely Deep Supervised Learning-based Computation Offloading, due to huge data transaction it consumes lots of energy. Hence, for saving energy through offloading, we use a multi-layer neural network as an ML tool. To achieve these goals our model contains three base matrices namely: 1) Task Computation; 2) Energy Consumption; and 3) Error Calculation.
Building: TASME Center
Room: Technology Hall
Date: July 4, 2021 - 11:35 AM – 11:50 AM