Concise pytorch implements of DRL algorithms, including REINFORCE, A2C, DQN, PPO(discrete and continuous), DDPG, TD3, SAC.

A clean and robust Pytorch implementation of PPO on continuous action space.

A clean and minimal implementation of PPO (Proximal Policy Optimization) algorithm in Pytorch, for continuous action spaces.

Pytorch Implement DRL algorithms (A2C, DDPG, PPO, TD3, SAC) for continuous action space control tasks.

Python + PyTorch. Advanced Reinforcement Learning (SAC/PPO/A2C) for ✨autonomous Robot Sumo combat featuring competitive self-play in continuous action spaces.

Implementation of Proximal Policy Optimization (PPO) for continuous action space (`Pendulum-v1` from gym) using tensorflow2.x and pytorch.

PPO pytorch tutorial for continuous control (BipedalWalker-v3)

Continuous asynchronous PPO in Pytorch solving OpenAI's bipdedal walker

The Second Project of Udacity Deep Reinforcement Learning Nano Degree. PyTorch implementation of PPO to solve Reacher Environment (Unity) with Continuous Action Space.

Pytorch implementation of Proximal Policy Optimization (PPO) for continuous action spaces

No description available

Continuous-control rocket landing environment and PPO agent in PyTorch.

PyTorch implementation of Proximal Policy Optimization (PPO) for discrete and continuous environments.

PyTorch PPO (discrete & continuous): clipped-surrogate actor, critic with TD-based advantage, minibatch updates & evaluation.

PPO Implementation from scratch, using Pytorch, Gymnasium, Numpy, and the Lunar Lander Continuous environment for testing both FF and RNN Policies

PyTorch implementation of chart-consistent mixture-of-experts PPO (CCMoE-PPO) for HalfCheetah-v5, including single-Gaussian PPO, MoE, and graph-Laplacian baselines, with training scripts, AUC-based evaluation, and paper-style figures for the associated CCMoE continuous-control study.

Proximal Policy Optimization (PPO) in PyTorch for continuous control in MuJoCo (DM Control Walker). Includes GAE, clipped objective, value function learning, and performance visualization.

Implemented and benchmarked PPO agent across pytorch, OpenAI Gym environments (CartPole, LunarLander, MountainCar) — studying policy gradient convergence, reward shaping, and hyperparameter sensitivity under continuous and discrete action spaces

A clean PyTorch implementation of PPO, SAC, and TD3 made from scratch. It is built for testing and comparing continuous control RL algorithms on complex environments such as BipedalWalker-v3.

Vision-based autonomous driving using Proximal Policy Optimization (PPO) in the CarRacing-v2 Gymnasium environment. The agent learns continuous control actions (steering, gas, brake) from raw pixel inputs using a CNN-based Actor-Critic architecture implemented with Stable-Baselines3 and PyTorch.

From-scratch implementation of Proximal Policy Optimization (PPO) for continuous control using the Pendulum-v1 environment. The project focuses on policy/value network design, numerical stability, hyperparameter sensitivity, and performance analysis. Implemented in PyTorch without high-level reinforcement learning frameworks.