ABSTRACT

With the advent of artificial intelligence (AI), there has been an attempt to model a similar behavior among computers, so that they can make decisions autonomously, just like humans. A significant subcategory of AI is machine learning (ML), where given some input, a machine can learn on its own from the available data and make accurate decisions or predictions on unseen data with the help of algorithms. When these learning algorithms are based on the neuron architectures of the CNS, this further categorizes ML into deep learning (DL). Combining the powers of DL and Internet of Things (IoT), the speed and efficiency with which complex tasks are accomplished across many connected computers are unmatched. Lack of sufficient resources is a major drawback in IoT devices, due to energy constraints and low computation capability when trying to deploy DL architectures for various applications. DL-based IoT is especially becoming popular in industries such as infrastructure, social networks, and content delivery.