ACM Home Page
Please provide us with feedback. Feedback
Concurrent imitation dynamics in congestion games
Full text PdfPdf (408 KB)
Source
Annual ACM Symposium on Principles of Distributed Computing archive
Proceedings of the 28th ACM symposium on Principles of distributed computing table of contents
Calgary, AB, Canada
SESSION: R2 table of contents
Pages 63-72  
Year of Publication: 2009
ISBN:978-1-60558-396-9
Authors
Heiner Ackermann  RWTH Aachen University, Aachen, Germany
Petra Berenbrink  Simon Fraser University, Burnaby, BC, Canada
Simon Fischer  RWTH Aachen University, Aachen, Germany
Martin Hoefer  RWTH Aachen University, Aachen, Germany
Sponsors
SIGOPS: ACM Special Interest Group on Operating Systems
SIGACT: ACM Special Interest Group on Algorithms and Computation Theory
ACM: Association for Computing Machinery
Publisher
ACM  New York, NY, USA
Bibliometrics
Downloads (6 Weeks): 18,   Downloads (12 Months): 35,   Citation Count: 1
Additional Information:

abstract   references   cited by   index terms   collaborative colleagues  

Tools and Actions: Request Permissions Request Permissions    Review this Article  
DOI Bookmark: Use this link to bookmark this Article: http://doi.acm.org/10.1145/1582716.1582732
What is a DOI?

ABSTRACT

Imitating successful behavior is a natural and frequently applied approach when facing scenarios for which we have little or no experience upon which we can base our decision. In this paper, we consider such behavior in atomic congestion games. We propose to study concurrent imitation dynamics that emerge when each player samples another player and possibly imitates this agents' strategy if the anticipated latency gain is sufficiently large. Our main focus is on convergence properties. Using a potential function argument, we show that these dynamics converge in a monotonic fashion to stable states. In such a state none of the players can improve their latency by imitating others.

As our main result, we show rapid convergence to approximate equilibria. At an approximate equilibrium only a small fraction of agents sustains a latency significantly above or below average. In particular, imitation dynamics behave like fully polynomial time approximation schemes (FPTAS). Fixing all other parameters, the convergence time depends only in a logarithmic fashion on the number of agents.

Since imitation processes are not innovative they cannot discover unused strategies. Furthermore, strategies may become extinct with non-zero probability. For the case of singleton games, we show that the probability of this event occurring is negligible. Additionally, we prove that the social cost of a stable state reached by our dynamics is not much worse than an optimal state in singleton congestion games with linear latency functions.

While we concentrate on the case of symmetric network congestion games, most of our results do not explicitly use the network structure. They continue to hold accordingly for general symmetric and asymmetric congestion games when each player samples within his commodity.


REFERENCES

Note: OCR errors may be found in this Reference List extracted from the full text article. ACM has opted to expose the complete List rather than only correct and linked references.

1
2
3
4
 
5
 
6
P. Berenbrink, T. Friedetzky, I. Hajirasouliha, and Z. Hu. Convergence to equilibria in distributed, selfish reallocation processes with weighted tasks. In Proc. 15th European Symposium on Algorithms (ESA), pages 41--52, 2007.
7
 
8
9
10
 
11
12
 
13
 
14
S. Fischer, L. Olbrich, and B. Vöcking. Approximating Wardrop equilibria with finitely many agents. Distributed Computing, 21(2):129--139, 2008.
15
 
16
 
17
18
 
19
 
20
J. Hofbauer and K. Sigmund. Evolutionary Games and Population Dynamics. Cambridge University Press, 1998.
 
21
S. Ieong, R. McGrew, E. Nudelman, Y. Shoham, and Q. Sun. Fast and compact: A simple class of congestion games. In Proc. 20th Conference on Artificial Intelligence (AAAI), pages 489--494, 2005.
 
22
E. Koutsoupias and C. Papadimitriou. Worst-case equilibria. In Proc. 16th Symposium on Theoretical Aspects of Computer Science (STACS), pages 404--413. Springer-Verlag, 1999.
 
23
 
24
R. Rosenthal. A class of games possessing pure-strategy Nash equilibria. International Journal of Game Theory, 2:65--67, 1973.
25
26
 
27
J. Weibull. Evolutionary Game Theory. MIT press, 1995.


Collaborative Colleagues:
Heiner Ackermann: colleagues
Petra Berenbrink: colleagues
Simon Fischer: colleagues
Martin Hoefer: colleagues