Finite Markov Decision Processes. This problem involves evaluative feedback and associative aspect. Refer to Chapter 3 Finite Markov Decision Processes by Sutton and Barto.

Dive into the heart of our RL, where a spectrum of exploration vs. exploitation algorithms awaits to enhance your decision-making prowess. Uncover the strategic brilliance of the Upper Confidence Bound (UCB) algorithm, dynamically balancing the quest for knowledge with a calibrated confidence-driven approach. Embrace simplicity and effectiveness with the Epsilon-Greedy algorithm, allowing you to seamlessly toggle between exploration and exploitation. For those navigating continuous action spaces, explore Gradient-Based Methods, leveraging derivatives for optimal search. Lastly, adopt the Bayesian perspective with Thompson Sampling, maintaining a distribution over possible models for principled decision-making in uncertain terrains. This repository offers a versatile suite of algorithms, each tailored to specific needs, providing you with the tools to elevate your exploration and exploitation strategies across diverse applications.

This is an app for Developing Data Products Class

Maximum likelihood estimation method, kernel density estimation and interactive graphs created with plotly