Reinforce-CartPole-v1

🎯 Reinforce Agent on CartPole-v1

This repository contains a trained Reinforce (Policy Gradient) agent that successfully solves the CartPole-v1 environment.

---

📊 Model Card

Model Name: Reinforce-CartPole-v1
Environment: CartPole-v1
Algorithm: Reinforce (Monte Carlo Policy Gradient)
Performance Metric:

• Achieves stable balancing of the pole across evaluation runs
• Mean reward approaches the environment’s maximum threshold

---

🚀 Usage

from huggingface_hub import load_from_hub
import gym

# Load the trained Reinforce model
model = load_from_hub(
    repo_id="KraTUZen/Reinforce-CartPole-v1",
    filename="reinforce.pkl"
)

# Initialize environment
env = gym.make(model["env_id"])

---

🧠 Notes

• The agent is trained using the Reinforce algorithm, which updates policy parameters via Monte Carlo returns.
• The environment is CartPole-v1, where the objective is to keep the pole balanced by moving the cart left or right.
• The serialized policy is stored in reinforce.pkl.

---

📂 Repository Structure

• reinforce.pkl → Trained policy weights
• README.md → Documentation and usage guide

---

✅ Results

• The agent consistently balances the pole for extended episodes.
• Demonstrates convergence to an optimal policy using policy gradient methods.

---

🔎 Environment Overview

• Observation Space: Continuous (cart position, velocity, pole angle, angular velocity)
• Action Space: Discrete (move cart left or right)
• Objective: Prevent the pole from falling by applying forces to the cart
• Reward: +1 for each timestep the pole remains upright

---

📚 Learning Highlights

• Algorithm: Reinforce (Policy Gradient)
• Update Rule: Policy parameters updated using returns from sampled episodes
• Strengths: Simple yet effective baseline for policy gradient methods
• Limitations: High variance in updates, mitigated with sufficient training episodes