Downloads · 30 days
0
dueling_dqn_cartpole
heisenberg5869/dueling_dqn_cartpole
dueling_dqn_cartpole is a reinforcement learning model from heisenberg5869. Use it for the reinforcement learning task on the model card, and read the license before you ship it in a product. It is set up for custom.
This repository contains a Dueling Deep Q-Network (Dueling DQN) agent trained on the CartPole-v2 environment. This model was implemented from scratch using PyTorch as part of an RL competition.
Access
Public
Updated Dec 29, 2025
Repo size
—
Likes
0
Public
Repo makeup
Click a slice to open those files.
.ipynb126 KB · 93%
At a glance
- Task
- Reinforcement Learning
- Library
- custom
- Access
- Public
- Created
- Dec 27, 2025
- Updated
- Dec 29, 2025
- SHA
- d30ad1d6
Model card
From the Hugging Face model README
Dueling DQN Agent for CartPole-v2
This repository contains a Dueling Deep Q-Network (Dueling DQN) agent trained on the CartPole-v2 environment. This model was implemented from scratch using PyTorch as part of an RL competition.
🚀 Performance
The agent is trained to balance the pole by applying forces to the cart.
- Algorithm: Dueling DQN (Separates State-Value and Advantage functions)
- Average Reward (100 episodes): 500.0 (Solved)
- Implementation: From scratch (no high-level RL libraries)
🛠️ Hyperparameters
{
"learning_rate": 1e-3,
"gamma": 0.99,
"epsilon_start": 1.0,
"epsilon_final": 0.0001,
"epsilon_decay_rate": 0.001,
"buffer_size": 10000,
"batch_size": 64,
"target_update_freq": 0.005
}
- Task
- Reinforcement Learning
- Library
- custom
- Created
- Dec 27, 2025
- Updated
- Dec 29, 2025