Bulletin Deccan College, Post Graduate & Research Institute, Deemed University A STUDY ON IMPLEMENTATION OF DEEP LEARNING TECHNIQUES IN DIFFERENT APPLICATIONS DURING COVID19 Prasanta Pratim Bairagi 1 ,Ankur Pan Saikia 2 ,Ananya Kalita 3 1,2 Computer Science & Engineering, Assam down town University 3 Assistant Professor, Civil Engineering, Assam down town University Abstract: Deep learning is an AI function that improves decision-making by simulating human data processing and pattern development. Deep learning is subtype of machine learning that employs a network to learn unsupervised from unstructured input. The epidemic of a new virus known as novelcoronavirus-2019 in China's Wuhan province has sounded the alarm for officials all over the world, as the number of persons affected and deaths has risen. To reduce the risk of infection, health officials recommend keeping a physical distance from others and wearing a face mask in public places. This research paper's objective is to analyze and explore the implementation of deep learning majorly, in preventing and implementing social norms prescribed by a world health organization to contain the spread of the novel corona virus. Keyword: Deep learning, Thermal screening, Face Recognition, Machine learning, Artificial intelligence Introduction Deep learning is a part of artificial intelligence that uses neural networks to learn from unstructured or unlabeled data. It is an artificial intelligence (AI) function that mimics the human brain's processing of data and creation of patterns for decision-making [1]. Deep learning techniques have accomplished brilliant success in different activities in computer vision, natural language processing, robotics and recently mitigating the novel-corona virus. Artificial intelligence (AI) refers to machinesthat have human-level or greater intellect that can abstract concepts from limited experience and transfer information across fields. Machine Learning is a modern Artificial Intelligence application based on the idea of giving machines access to data and allowing them to learn for them [Ganatra, Nilay & Patel, Atul. (2018)]. Face recognition accuracy and speed have improved dramatically because to advances in deep learning and the advent of deep convolution neural networks. Deep Learning is a subset of machine learning, which is an intern subset of Artificial Intelligence, a popular topic of computer science over the last decade, as shown in Figure 1. There are three types of learning: supervised, semi- supervised, and unsupervised. Deep learning refers to a set of techniques and methods for learning features and tasks from data. Data can be structured or unstructured, and it might include images, text, or sound. Because it learns directly from data, deep learning is commonly referred to as end-to-end learning. Deep learning algorithms, on the other hand, work without the intervention of a human and are sometimes capable of delivering more accurate findings than a human [Yang Li, Sangwhan]. 1 Assistant Professor Assam down town University 2 Corresponding Author