Q-Strategy: Automated Bidding and Convergence in Computational Markets

摘要

Agents and market mechanisms are widely elaborated and applied to automate interaction and decision pro-cesses among others in robotics, for decentralized con-trol in sensor networks and by algorithmic traders in fi-nancial markets. Currently there is a high demand of ef-ficient mechanisms for the provisioning, usage and allo-cation of distributed services in the Cloud. Such mech-anisms and processes are not manually manageable and require decisions made in quasi real-time. Thus agent decisions should automatically adapt to changing con-ditions and converge to optimal values. This paper presents a bidding strategy, which is capa-ble of automating the bid generation and utility maxi-mization processes of consumers and providers by the interaction with markets as well as to converge to opti-mal values. The bidding strategy is applied to the con-sumer side against benchmark bidding strategies and its behavior and convergence are evaluated in two market mechanisms, a centralized and a decentralized one.