Abstract

We approach the privacy problem as one of information revelation in a multi-stage game. The optimal play in one stage reveals information about future possible plays, which can be used to the player's advantage or disadvantage. An example is eBay's second chance offer, where information from a losing bid is used to price discriminate against a bidder. We present two sets of results for a generalization of the second-chance offer. The first demonstrates how game-theoretic approaches to privacy protection can be more useful to the bidder than cryptographic approaches. The second set presents optimal bidding strategies when the bidder believes the seller will price discriminate; showing that if the probability of price discrimination is high, the bidder quantizes his bid, thus reducing the amount of information revealed about the maximum he is willing to pay for the object. This is joint work with economics faculty member Sumit Joshi and computer science doctoral student Yu-An Sun.

Knowledge of game theory will not be assumed, and the basic results from auction theory required will be covered in the talk.

Speaker Biography

Poorvi L. Vora is an Assistant Professor in the Department of Computer Science at The George Washington University. Her current interests are in privacy, electronic voting and cryptology. Before GW, she worked at Hewlett-Packard -- in HP Labs and in the product business -- on digital cameras, counterfeit deterrence, watermarking, DRM and privacy.