Inspired by the gym environments from POCAR
A Gym environment that simulates the process of college admissions, particularly exploring the relationship between economic burden and students' acceptance. This environment provides fairness metrics and generates plots to assess the performance of different RL-based decision-making algorithms in a simulation, allowing concrete baseline assessment of their long-term fairness.
Install Anaconda to setup a virtual environment.
Run the following commands
conda create -n college python=3.8
conda activate college
pip install -r requirements.txt
Usage:
Navigate to the directory where main.py is located and then run the following:
python main.py
This will train two agents - A-PPO and G-PPO and evaluate their fairness and utility performance in our environment, outputting graphs. Follow this tutorial for adding/running more RL decision-making models.
Can change training time or episode length in config.py
College_Admissions_MDP_Cycle.png provides a visualization of how the environment works.