Autonomous Bidding Agents in the Trading Agents Competition

A natural offshoot of the growing prevalence of online auctions is the creation of autonomous bidding agents that monitor and participate in these auctions. It is straightforward to write a bidding agent to participate in an online auction for a single good, particularly when the value of that good is fixed ahead of time: the agent can bid slightly over the ask price until the auction closes or the price exceeds the value. In simultaneous auctions offering complementary and substitutable goods, however, agent deployment is a much more complex endeavor. The first trading agent competition (TAC), held in Boston, Massachusetts, on 8 July 2000, challenged participants to design a trading agent capable of bidding in online simultaneous auctions for complimentary and substitutable goods. TAC was organized by a group of researchers and developers led by Michael Wellman of the University of Michigan and Peter Wurman of North Carolina State University. In a companion article, the tournament organizers present the design and operation of the competition.1 This article describes the task-specific details of, and the general motivations behind, the four top-scoring agents. First, we discuss general strategies used by most of the participating agents. We then report on the strategies of the four top-placing agents. We conclude with suggestions for improving the design of future trading agent competitions.