Manisha2612 / Deep-Reinforcement-Learning-Implementation

Deep Reinforcement Learning Project

EE-556-Final-Project

EE 556 Final Project

In this project, we have implemented deep reinforcement learning using DQN and DDQN agents with and without Priortized Experience replay on several agents. In second part of project, we have implemented Quantile regression Distribution learning and observed the rewards gained by environments

Convolution Neural Network:

DQN, DDQN, PER results for CartPole-v0 and LunarLander-v2

Super Mario Results:

Quantile Regression Results

CartPole and LunarLander

Super Mario Bros