Paper Title
Comparative Study of Reinforcement Learning Algorithms and Their Applications in Healthcare Sector

Abstract
Expansion in diseases troubled the country’s well-being and financial state. Covid-19 upshot a lot of consequences for mankind, lead the way to make more efforts to detect and put a stop to its spread. No one can abolish the diseases utterly but still the timely prediction of the diseases set free the life of very many patients. Reinforcement Learning, the most trending branch of Machine learning put into practice in the healthcare sector for diagnosing copious diseases at the commencing stage. This approach is somehow different in such a way that it does not require anterior knowledge rather the knowledge is acquired by continuously interacting with the environment. Reinforcement Learning works on the development of autonomous systems with the help of trained agents who interacts with the environment to find out the optimal behaviour. Reinforcement Learning proves to be much beneficial in the healthcare sector which is the main objective of this paper. This paper investigates various algorithms used in reinforcement learning like Q-Learning, Sarsa algorithms, Monte-Carlo methods, Deep Q-Learning and further Deep learning approaches like CNN, RNN, LSTM, ANN, Bi-LSTM, DDPG etc. Also, the paper works for the healthcare applications wherein many diverse area of diseases are considered for analysis like heart disease, Lung disease, Liver disease, Diabetes, malaria, Monkey-pox and many more like this and it is observed that which algorithm proves to be the finest. Finally, after investigating the functioning of these algorithms for several diseases, come to the conclusion that CNN bring about the finest. Keywords - Deep Deterministic Policy Gradient (DDPG), Periapical Radiographic Images (PRI), Neural Networks, Bidirectional- Long Short Term Memory (Bi-LSTM), Gastro-Intestinal Bleeding Events.