2048 environment for Reinforcement Learning and DQN algorithm

Deep Reinforcement Learning with Double Q-learning

Deep Reinforcement Learning based Decision-Making in Autonomous Driving Tasks

This is an implementation of Deep Reinforcement Learning for a navigation task. Specifically, DQN algorithm with experience replay method is used to solve the task.

This repo hosts a sophisticated reinforcement learning setup for training a DQN agent in “CarRacing-v2”. It has self-adaptive features like dynamic learning rate and domain randomization to boost agent training and performance. It includes an Evaluation Callback for optimal model retention and leverages GPU for quicker training.

First I created an environment of openAI and Gymnasium I have campared Q-Learning Algoirthm and and DQN Learning Algorithm I got best reward DQN Because It's advance

Simple breakout game with DQN agent which learn how to play it.

a 2D platformer game made with Unity engine and C#

The "Reinforcement Learning Snake Game" project uses Deep Q-Learning to train an AI agent to play Snake autonomously. The agent learns to maximize its score by eating apples and avoiding collisions, demonstrating reinforcement learning in a game environment. The project includes game logic, RL agent code, and training scripts.

Reinforcement Learning: Cartpole Balancing with a DQN Agent

Implementations of some of the most well known Deep Reinforcement Learning algorithms

This repository contains a comprehensive implementation of a Deep Q-Network (DQN) to train an AI agent to play Atari's Breakout game. The implementation leverages OpenAI Gym for the game environment and TensorFlow/Keras for the neural network. Features include experience replay, target networks, and game monitoring via exported videos.