Return of the boss problem: competing online against a non-adaptive adversary

Authors:
Magnús M. Halldórsson;Hadas Shachnai
Affiliations:
School of Computer Science, Reykjavik University, Reykjavik, Iceland;Department of Computer Science, The Technion, Haifa, Israel
Venue:
FUN'10 Proceedings of the 5th international conference on Fun with algorithms
Year:
2010

Citing 11
Cited 0

On the performance of on-line algorithms for partition problems

Acta Cybernetica
A better lower bound for on-line scheduling

Information Processing Letters
Better bounds for online scheduling

STOC '97 Proceedings of the twenty-ninth annual ACM symposium on Theory of computing
Online computation and competitive analysis

Online computation and competitive analysis
Robust Matchings

SIAM Journal on Discrete Mathematics
All-norm approximation algorithms

Journal of Algorithms
Convex programming for scheduling unrelated parallel machines

Proceedings of the thirty-seventh annual ACM symposium on Theory of computing
Robbing the bandit: less regret in online geometric optimization against an adaptive adversary

SODA '06 Proceedings of the seventeenth annual ACM-SIAM symposium on Discrete algorithm
Improved results for data migration and open shop scheduling

ACM Transactions on Algorithms (TALG)
Robust cost colorings

Proceedings of the nineteenth annual ACM-SIAM symposium on Discrete algorithms
On the max coloring problem

WAOA'07 Proceedings of the 5th international conference on Approximation and online algorithms

Quantified Score

Hi-index	0.00

Visualization

Abstract

We follow the travails of Enzo the baker, Orsino the oven man, and Beppe the planner. Their situation have a common theme: They know the input, in the form of a sequence of items, and they are not computationally constrained. Their issue is that they don't know in advance the time of reckoning, i.e. when their boss might show up, when they will be measured in terms of their progress on the prefix of the input sequence seen so far. Their goal is therefore to find a particular solution whose size on any prefix of the known input sequence is within best possible performance guarantees.