Algorithmics – is there hope for a unified theory?
CSR'10 Proceedings of the 5th international conference on Computer Science: theory and Applications
Hi-index | 0.00 |
Given the diversity of computational problems, models of computation, solution concepts, and algorithmic approaches, it often seems a challenge to view computer science/informatik as a coherent discipline. Fortunately, various unifying ideas (e.g. NP-completeness) have brought significant perspective to a very complex and ever expanding field.In a less ambitious program, I have been thinking about simple algorithmic approaches that arise frequently in approximation algorithms for combinatorial optimization. To be more specific, I am in interested in "greedy-like" algorithms and recently I am trying to understand the use of such algorithms in combinatorial auctions. I will discuss some precise models for simple greedy-like algorithms and how such algorithms can and cannot achieve desirable properties such a truthfulness, and convergence to "good" equilibria.The ideas in this talk are based on recent work with Brendan Lucier and also relate to Yuli Ye's talk in ICALP.