An adaptive bidding strategy for combinatorial auctions-based resource allocation in dynamic markets

Combinatorial auctions, where bidders are allowed to submit bids for bundles of items, are preferred to single-item auctions when bidders have complementarities and substitutabilities among items and therefore achieve better social efficiency. A large unexplored area of research is the design of bidding strategies. In this paper, we propose a new adaptive bidding strategy for combinatorial auctions-based resource allocation. The bidder adopting this strategy can adjust his profit margin constantly, and thus perceive and respond to the market in a timely way. Experiment results show that the adaptive strategy performs fairly well when compared to an intelligent strategy, which is artificially generated according to prior knowledge of the market.