This repository contains solutions for the Frozen Lake problem using PEARL, as well as implementations of Contextual Bandit using PEARL and Neural Bandit with LinUCB, SquareCB, and LinTS. The project is part of the coursework for the Machine Learning course at the University of Kashan.

• Objective: Solve the Frozen Lake game, a popular reinforcement learning problem, utilizing the Probabilistic Embeddings for Actor-critic Reinforcement Learning (PEARL) algorithm.

• Objective: Implement a Contextual Bandit solution using the PEARL algorithm to dynamically adjust and optimize decision-making strategies based on contextual information.

• Objective: Explore the application of Neural Bandits through the implementation of various algorithms including LinUCB, SquareCB, and LinTS, focusing on their performance and efficiency in decision-making processes.