Deep Learning-Based Dew Computing with Novel Offloading Strategy

Deep learning applications are prevalent. Its popularity is increasing day by day. But the deep learning model cannot be efficiently run with any device. If we want to take advantage of this up to low-level devices, we have to find a unique way. Dew Computing (DC) has arisen as a modern computational paradigm, Wide Cloud Storage acceptability. This paper has shown how to use an offloading strategy and use the dew computing layer to efficiently run a deep learning application without the internet at low latency. Moreover, we also showed how a deep learning model could have an online impact when it comes to training and how long it takes to train.