Upper Confidence Bound based Decision Making Strategies and Dynamic Spectrum Access
The UCB algorithms aims to solve the current problems faced in channel estimation schemes. The dilemma faced during channel estimation is mostly due to the incomplete information provided before accessing a particular channel. The concept and basis of Reinforcement Learning is what makes the Cognitive Agent better by deciding based on the previous interactions.
The UCB.py file present in the folder contains the implemented Python Code for the same which uses the probability concept and UCB formulations to train the cognitive agent first and take better decisions later in order to select the best suitable Channel for further estimations.