README.md

RL Learn

I'm implementing algorithms from this book (Mathematical Foundation of Reinforcement Learning). I used the GridWorld environment code provided by the author and made a few changes.

Menu

• Chapter 4
  • Value Iteration
  • Policy Iteration
  • Truncated Policy Iteration
• Chapter 5
  • MC Basic
  • MC Exploring Starts
  • MC $\epsilon$-greedy
• Chapter 6
  • RM Algorithm example
• Chapter 7
  • Sarsa
  • Q-learning (on-policy version)
  • Q-learning (off-policy version)

Run

It is easy to run the code by following the steps.

1. activate your venv

conda activate base
# or
python -m venv venv
. venv/bin/activate

2. install packages

pip install -r requirements.txt

3. clone this repo

git clone [email protected]:co-gy/rl-learn.git

4. run

make sure your running directory is rl-learn

cd rl-learn
python algorithm/value_iteration.py