This repository provides PyTorch implementations of four deep reinforcement learning algorithms: Advantage Actor Critic (A2C), Proximal Policy Optimization (PPO), ACKTR using Kronecker-factored approximation, and Generative Adversarial Imitation Learning (GAIL). The implementation supports Atari, MuJoCo, PyBullet, and Roboschool environments via OpenAI Gym. It serves as a research-grade RL training codebase inspired by OpenAI baselines.