In this repo, I train Q-Learning based agent to navigate the large square world and collect bananas

Train an agent using RL to navigate (and collect bananas) in a large, square world

This repository contains code related to "Project 1 - Navigation" from Udacity's Deep Reinforcement Learning Nanodegree program.

Keeping the double-jointed arm hand in the green sphere

Training a pair of agents to play tennis

Learning Continuous Control in Deep Reinforcement Learning

Code to train a RL agent to solve the Reacher environment [Unity ML-Agents]

Train a system of DeepRL agents to demonstrate collaboration or cooperation on a complex task.

In this repo, I implement deep deterministic policy gradients and multi-agent deep deterministic poilicy gradients to solve the Tennis enironment (Unity ML-Agents)

Proximal Policy Optimization with Beta distribution - uses multi agent Unity ML Tennis

Project 3 of Udacity Deep Reinforcement Learning Nanodegree

Implementation of the DDPG algorithm to solve Continuous Control Reacher Environment

Deep Reinforcement Learning: Two Competing Tennis Playing Agents

Continuous Control with deep reinforcement learning where the agent must reach a moving ball with a double jointed arm

Deep Reinforcement Learning: Navigation. Capture yellow bananas while avoiding blue bananas with an agent trained using Deep Q-Networks on a Unity ML-Agent environment.

Continuous Control Project for Deep Reinforcement Learning Nanodegree

Navigating a banana world using Dueling Double DQN network

Training a pair of competing RL agents to play Tennis using MADDPG algorithm

AI based Massive Multiplayer Online Simulation