MLP-framework (pure numpy) and DDQN-framework for OpenAI's Gym games. +test code for PPO added. +Hindsight Experience Replay(HER) bitflip-DQN example. +prioritized replay.

Official implementation of the paper "Approximating two value functions instead of one: towards characterizing a new family of Deep Reinforcement Learning Algorithms": https://arxiv.org/abs/1909.01779 To appear at the next NeurIPS2019 DRL-Workshop

Quadcopter Controller - Reinforcement Learning Project for Udacity Machine Learning Engineer Nanodegree

This repository contains the reinforcement learning framework for training a 2D car agent to navigate dynamically evolving obstacle courses, developed as a final project for the Reinforcement Learning course (academic year 24/25).

Deep Reinforcement Learning methods on Lunar Lander from OpenAI Gym.