Reinforcement Learning Algorithms

This repository contains PyTorch implementations of three policy gradient algorithms, designed for continuous control tasks and tested primarily on the HalfCheetah-v4 environment.

Implemented Algorithms

DDPG (Deep Deterministic Policy Gradient): Off-policy algorithm combining DQN and policy gradients for continuous action spaces
SAC (Soft Actor-Critic): Off-policy algorithm with entropy regularization for improved exploration
PPO (Proximal Policy Optimization): On-policy algorithm with clipped surrogate objective for stable updates