This repository contains code of numerical experiments for paper:
Peer Prediction for Learning Agents
NeurIPS 2022
Shi Feng, Fang-Yi Yu, Yiling Chen
[ArXiv Version]
If you want to draw convergence rates of learning algorithms in CA mechanism, you need to run
python main.py --converge_rate
If you want to draw error bars for convergence rates of learning algorithms in CA mechanism, you need to run
python main.py --error_bar
Here is one of our experiment results:
.
├── game/ # components in a peer prediction game
│ ├── agent.py # implementing agents
│ └── game.py # implementing signal generator and CA mechanism
├── utils/ # drawing simulation results
│ ├── converge_rate.py
│ └── error_bar.py
├── results/ # our running samples
├── main.py # main file
├── .gitignore # exclude some annoying files from git
├── LICENSE # MIT license
└── README.md # what you are reading now
If you have any questions, feel free to contact us through email ([email protected]) or Github issues. Enjoy!