Benchmark for Auto-bidding Task
from Neural intel Pod Hosted by Neuralintel.org
This document introduces BAT (Benchmark for Auto-bidding Task), a new resource for researching autobidding algorithms in online advertising auctions. It provides a large-scale dataset from the Avito platform, including information on two common auction types, VCG and FP, along with campaign details, auction outcomes, and traffic patterns. The paper also presents several baseline autobidding algorithms tested on this dataset and evaluates their performance on key challenges like budget pacing uniformity and Cost Per Click (CPC) constraint optimization. The resource aims to offer a standardized and user-friendly framework for researchers to develop and assess new automated bidding strategies.
Embed this episode
No transcript yet.