Theoretical Improvements in Algorithmic Efficiency for Network Flow Problems
Journal of the ACM (JACM)
AdWords and Generalized On-line Matching
FOCS '05 Proceedings of the 46th Annual IEEE Symposium on Foundations of Computer Science
Budgeted Allocations in the Full-Information Setting
APPROX '08 / RANDOM '08 Proceedings of the 11th international workshop, APPROX 2008, and 12th international workshop, RANDOM 2008 on Approximation, Randomization and Combinatorial Optimization: Algorithms and Techniques
FOCS '08 Proceedings of the 2008 49th Annual IEEE Symposium on Foundations of Computer Science
The adwords problem: online keyword matching with budgeted bidders under random permutations
Proceedings of the 10th ACM conference on Electronic commerce
Online primal-dual algorithms for maximizing ad-auctions revenue
ESA'07 Proceedings of the 15th annual European conference on Algorithms
| Hi-index | 0.00 |
Given a bipartite graph G=(U,V,E) with U={1,…,n}, and a positive budget Bv for each v in V, a B-matching M in G is a second-price B-matching if, for every edge uv in M, there is a uw in E so that less than Bw edges u′w with u′u belong to M. The Second-Price Ad Auction with Binary Bids (B2PAA) consists of, given G and B as above, finding a second-price B-matching in G as large as possible. The particular case of this problem where Bv=1 for all v, denoted as Second-Price Matching (2PM), is known to be APX-hard and there is a 2-approximation for it. We present a way to use this approximation and similar ones to approximate B2PAA. Also, we formalize the idea of a competitive market, present an improved approximation for 2PM on competitive markets, extend the inapproximability results for competitive markets and analyze the performance of an algorithm of Azar, Birnbaum, Karlin, and Nguyen for the online 2PM on competitive markets. Our improved approximation can also be used for B2PAA. Finally, we comment on results derived from computational experiments on variants of our algorithm.