TREE CONFIGURATION GAMES FOR DISTRIBUTED STREAM MINING SYSTEMS

We consider the problem of configuring classifier trees in dis-tributed stream mining system. The configuration involves selecting appropriate false-alarm detection tradeoffs for each classifier to minimize end-to-end penalty in terms of misclas-sification cost. We model this as a tree configuration game and design solutions, where individual classifiers select their operating points to maximize a local utility. We derive ap-propriate misclassification cost coefficients for intermediate classifiers, and determine the information that needs to be ex-changed across classifiers, in order to successfully design the game. We analytically show that there is a unique pure strat-egy Nash equilibrium in operating points, which guarantees a convergence of the proposed approach. We evaluate the per-formance of our algorithm on an application for sports scene classification, and compare against centralized solutions. We show that our algorithm results in better performance than the centralized solution on average. Moreover, the algorithm ap-proaches the optimal solution asymptotically with increasing number of actions per classifier. Index Terms — Resource constrained stream mining, tree configuration games, binary classifier tree. 1