We show that any communication finding a Pareto efficient allocation in a private-information economy must also discover supporting Lindahl prices. In particular, efficient allocation of L indivisible objects requires naming a price for each of the 2 L ¡1 bundles. Furthermore, exponential communication in L is needed just to ensure a higher share of surplus than that realized by auctioning all items as a bundle, or even a higher expected surplus (for some probability distribution over valuations). When the valuations are submodular, efficiency still requires exponential communication (and fully polynomial approximation is impossible). When the objects are homogeneous, arbitrarily good approximation is obtained using exponentially less communication than that needed for exact efficiency.

Combinatorial auctions (CAs) where bidders can bid on bundles of items can be very desirable market mechanisms when the items sold exhibit complementarity and/or substitutability, so the bidder's valuations for bundles are not additive. However, in a basic CA, the bidders may need to bid on exponentially many bundles, leading to di#culties in determining those valuations, undesirable information revelation, and unnecessary communication. In this paper we present a design of an auctioneer agent that uses topological structure inherent in the problem to reduce the amount of information that it needs from the bidders. An analysis tool is presented as well as data structures for storing and optimally assimilating the information received from the bidders. Using this information, the agent then narrows down the set of desirable (welfare-maximizing or Pareto-e#cient) allocations, and decides which questions to ask next. Several algorithms are presented that ask the bidders for value, order, and rank information. A method is presented for making the elicitor incentive compatible.

This paper focuses on load management in looselycoupled federated distributed systems. We present a distributed mechanism for moving load between autonomous participants using bilateral contracts that are negotiated offline and that set bounded prices for moving load. We show that our mechanism has good incentive properties, efficiently redistributes excess load, and has a low overhead in practice.

We consider the parallels between the preference elicitation problem in combinatorial auctions and the problem of learning an unknown function from learning theory. We show that learning algorithms can be used as a basis for preference elicitation algorithms. The resulting elicitation algorithms perform a polynomial number of queries. We also give conditions under which the resulting algorithms have polynomial communication. Our conversion procedure allows us to generate combinatorial auction protocols from learning algorithms for polynomials, monotone DNF, and linear-threshold functions. In particular, we obtain an algorithm that elicits XOR bids with polynomial communication. We then characterize the communication requirements of implementing Vickrey payments with an elicitation algorithm. This suggests a modification to the queries in our elicitation algorithms so that truthful bidding becomes an ex-post Nash equilibrium.

We consider auction design in a setting with costly preference elicitation. We motivate the role of proxy agents, that are situated between bidders and the auction, and maintain partial information about agent preferences and compute equilibrium bidding strategies based on the available information. The proxy agents can also elicit additional preference information incrementally during an auction. We show that indirect mechanisms, such as proxied ascending-price auctions, can achieve better allocative efficiency with less preference elicitation than direct mechanisms, such as sealed-bid auctions.

Combinatorial auctions allow bidders to express complex valuations on bundles of items, and have been proposed in settings as diverse as the allocation of floor space in a new condominium building to individual units (Wired 2000) and the allocation of take-off and landing slots at airports (Smith, Forward). Many applications are described in Part V of this book. The promise of combinatorial auctions (CAs) is that they can allow bidders to better express their private information about preferences for different outcomes and thus enhance competition and market efficiency. Much effort has been spent on developing algorithms for the hard problem of winner determination once bids have been received (Sandholm, Chapter 14). Yet, preference elicitation has emerged as perhaps the key bottleneck in the real-world deployment of combinatorial auctions. Advanced clearing algorithms are worthless if one cannot simplify the bidding problem facing bidders. Preference elicitation is a p

Abstract not found

A simple characterization of the equilibrium conditions required to compute Vickrey payments in the Combinatorial Allocation Problem leads to an ascending price Generalized Vickrey Auction. The ascending auction, /Bundle Extend & Adjust (,BEA), maintains non-linear and perhaps non-anonymous prices on bundles of items, and terminates with the efficient allocation and the Vickrey payments in ex post Nash equilibrium.

The paper examines the communication requirements of social choice rules when the (sincere) agents privately know their preferences. It shows that for a large class of choice rules, any communication verifying that an alternative is in the rule must reveal supporting budget sets for the agents such that the optimality of the proposed alternative to all agents within their respective budget set in itself verifies the al-ternative. We characterize the budget equilibria that are the minimally informative messages verifying a given choice rule. This characterization is used to identify the communication burden of choice rules, measured with the number of transmitted bits or real variables. Applications include efficiency in convex economies, exact or approximate surplus maximization in combinatorial auctions, the core in indivisible-good economies, and stable many-to-one matchings.

Abstract. In this paper, we suggest using an English Auction Protocol for a procurement multi-attribute auction in which the item for sale is defined by several attributes, the buyer agent is the auctioneer, and the seller agents are the bidders. Such domains include auctions on task allocation, services, or compound products. At the beginning of the auction the buyer agent announces the required properties of the item, and then various seller agents propose bids, which are composed of specific configurations that match its request. Each proposed bid should be better for the buyer agent than the previous bid, w.r.t. the announced requirements of the buyer agent. Finally, the last suggested bid will win, and the seller agent that suggested this bid will be committed to it. We consider two utility function models for the English auction protocols and provide the optimal bidding strategies for the seller agents and the optimal auction design for the buyer agents regarding both models. 1