amazon-science/auction-gym

AuctionGym is a simulation environment that enables reproducible evaluation of bandit and reinforcement learning methods for online advertising auctions.

57
/ 100
Established

This project helps advertising professionals, data scientists, and researchers evaluate new strategies for bidding in online ad auctions without the cost and risk of live experiments. You provide details about auction settings and bidder behaviors, and it simulates how different bidding and ad allocation methods perform, generating metrics and visualizations. This is for anyone who needs to test and understand the impact of various bidding algorithms in a controlled, reproducible environment.

187 stars.

Use this if you need a reliable way to benchmark novel ad bidding methods and understand their impact on advertiser welfare and auctioneer revenue before deploying them in real-world campaigns.

Not ideal if you are looking for a tool to manage live ad campaigns or optimize bids directly within an existing ad platform.

online-advertising ad-bidding-strategy auction-simulation marketing-analytics ad-tech-research
No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 21 / 25

How are scores calculated?

Stars

187

Forks

45

Language

Jupyter Notebook

License

Apache-2.0

Last pushed

Feb 11, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/amazon-science/auction-gym"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.