William E. Walsh

The Michigan Internet AuctionBot: a configurable auction server for human and software agents

Market mechanisms such as auctions will likely rep resent a common interaction medium for agents on the Internet The Michigan Internet AuctionBot is a ex ible scalable and robust auction server that supports both software and human agents The server manages many simultaneous auctions by separating the interface from the core auction procedures This design provides a responsive interface and tolerates system and network disruptions but necessitates careful timekeeping proce dures to ensure temporal accuracy The AuctionBot has been used extensively in classroom exercises and is available to the general Internet population Its exi ble speci cation of auctions in terms of orthogonal pa rameters makes it a useful device for agent researchers exploring the design space of auction mechanisms

Auction Protocols for Decentralized Scheduling

Scheduling is the problem of allocating resources to alternate possible uses over designated periods of time. Several have proposed (and some have tried) market-based approaches to decentralized versions of the problem, where the competing uses are represented by autonomous agents. Market mechanisms use prices derived through distributed bidding protocols to determine an allocation, and thus solve the scheduling problem. To analyze the behavior of market schemes, we formalize decentralized scheduling as a discrete resource allocation problem, and bring to bear some relevant economic concepts. Drawing on results from the literature, we discuss the existence of equilibrium prices for some general classes of scheduling problems, and the quality of equilibrium solutions. To remedy the potential nonexistence of price equilibria due to complementarities in preference, we introduce additional markets in combinations of basic goods. We present some auction mechanisms and bidding protocols corresponding to the two market structures, and analyze their computational and economic properties. Finally, we consider direct revelation mechanisms, and compare to the market-based approach.

Flexible double auctions for electionic commerce: theory and implementation

Abstract We consider a general family of auction mechanisms that admit multiple buyers and sellers, and determine market-clearing prices. We analyze the economic incentives facing participants in such auctions, demonstrating that, under some conditions, it is possible to induce truthful revelation of values by buyers or sellers, but not both, and for single- but not multi-unit bids. We also perform a computational analysis of the auctioneers task, exhibiting efficient algorithms for processing bids and calculating allocations.

A Parametrization of the Auction Design Space

Abstract We present an extensive breakdown of the auction design space that captures the essential similarities and differences of many auction mechanisms in a format more descriptive and useful than simple taxonomies. This parametrization serves as an organizational framework in which to classify work within the field and uncovers parameter combinations corresponding to novel mechanisms. The structured characterization of auction rules can be exploited for the modular design of configurable auction servers. It also facilitates the communication of auction rules to software agents, enabling the automation of flexible market-based negotiation. Journal of Economic Literature Classification Numbers: C70, D44.

Combinatorial auctions for supply chain formation

Supply chain formation presents difficult coordination issues for distributed negotiation protocols. Agents must simultaneously negotiate production relationships at multiple levels, with important interdependencies among inputs and outputs at each level. Combinatorial auctions address this problem by global optimization over expressed offers to engage in compound exchanges. A one-shot combinatorial auction that optimizes the reported value of the bids results in optimal allocations with truthful bids. But autonomous self-interested agents have an incentive to bid strategically in an attempt to gain extra surplus. We investigate a particular combinatorial protocol consisting of a one-shot auction and a strategic bidding policy. We experimentally analyze the efficiency and producer surplus obtained in five networks, and compare this performance to that of a distributed, progressive auction protocol with non-strategic bidding. We find that producers can sometimes gain significantly by bidding strategically. However, when the available surplus is small relative to the consumers’ values, the producers’ strategic behavior may prevent the supply chain from forming at all, resulting in zero gains for all agents. We examine the robustness of the combinatorial protocol by investigating agent incentives to deviate, identifying quasi-equilibrium behavior for an example network.

Designing the market game for a trading agent competition

The authors discuss the design and operation of a trading agent competition, focusing on the game structure and some of the key technical issues in running and playing the game. They also describe the competitions genesis, its technical infrastructure, and its organization. The article by A. Greenwald and P. Stone (2001), describes the competition from a participants perspective and describes the strategies of some of the top-placing agents. A visualization of the competition and a description of the preliminary and final rounds of the TAC are available in IC Online (http://computer.org/internet/tac.htm).

A market protocol for decentralized task allocation

We present a decentralized market protocol for allocating tasks among agents that contend for scarce resources. Agents trade tasks and resources at prices determined by an auction protocol. We specify a simple set of bidding policies that, along with the auction mechanism, exhibits desirable convergence properties. The system always reaches quiescence. If the system reaches quiescence below the consumers reserve price for the high level task, it will be in a solution state. If the system finds a solution it will reach quiescence in a solution state. Experimental evidence supports our conjecture that the system will converge to a solution when one exists and the consumer bids sufficiently high. We describe the systems application to and implementation in an agent-based digital library.

Some economics of market-based distributed scheduling

Market mechanisms solve distributed scheduling problems by allocating the scheduled resources according to market prices. We model distributed scheduling as a discrete resource allocation problem, and demonstrate the applicability of economic analysis to this framework. Drawing on results from the literature, we discuss the existence of equilibrium prices for some general classes of scheduling problems, and the quality of equilibrium solutions. We then present two auction protocols for implementing solutions, and analyze their computational and economic properties.

Modeling supply chain formation in multiagent systems

Supply chain formation is an important problem in the commercial world, and can be improved by greater automated support. Hence, the multiagent systems community should work to develop new solutions to the problem. The problem is complex and challenging, and a complete model must encompass a number of issues. In this paper we highlight some issues that must be understood to make progress in modeling supply chain formation.

Distributed quiescence detection in multiagent negotiation

In a distributed multiagent negotiation involving multiple issues, it is often desirable to finalize deals only when all related issues are resolved. However, detecting that a multiagent negotiation has reached a globally quiescent state can be a nontrivial task in a distributed, asynchronous system. We present a quiescence detection protocol based on the Dijkstra-Scholten algorithm for distributed termination detection (D.W. Dijkstra and C.S. Scholten, 1980). The protocol operates as a layer on top of an underlying mediated negotiation protocol. If agents conform to the detection protocol, the detection process terminates iff the negotiation is quiescent. We discuss agent incentives to deviate from the protocol, and describe extensions that enforce adherence with respect to the most significant potential deviations.