Artificial Intelligence for Multi-Unit Auction design

TL;DR

This paper explores how artificial intelligence, especially reinforcement learning, can be used to model and improve bidding strategies in multi-unit auctions, addressing a key challenge in auction theory.

Contribution

It introduces six AI algorithms for learning and bidding in multi-unit auctions and demonstrates their application through comparative analysis.

Findings

AI algorithms effectively simulate bidding behavior

Reinforcement learning enhances auction design insights

Comparison shows varied performance of algorithms

Abstract

Understanding bidding behavior in multi-unit auctions remains an ongoing challenge for researchers. Despite their widespread use, theoretical insights into the bidding behavior, revenue ranking, and efficiency of commonly used multi-unit auctions are limited. This paper utilizes artificial intelligence, specifically reinforcement learning, as a model free learning approach to simulate bidding in three prominent multi-unit auctions employed in practice. We introduce six algorithms that are suitable for learning and bidding in multi-unit auctions and compare them using an illustrative example. This paper underscores the significance of using artificial intelligence in auction design, particularly in enhancing the design of multi-unit auctions.