This project consists of three milestone projects

• Milestone 1: Training a two-layer neural network based on XOR gate truth table with the error-backpropagation algorithm. All the relevant codes and files are in the Backpropagation_Learning folder

• Milestone 2: Implementeing reinforcement learning for a robot tank in Robocode environment with a look-up table and comparing the reinforcement learning performance between on-policy and off-policy. All the relevant codes and files are in the Reinforcement_Learning_with_LUT folder

• Milestone 3: Replacing the look-up table with the neural network, which has the optimum hyperparameters, to train the robot tank in the Robocode environment. All the relevant codes and files are in the Reinforcement_Learning_with_Backpropagation folder