A real world network pricing game with less severe Braess' Paradox
Abhimanu Kumar, Sanjib Kumar Das

TL;DR
This paper introduces a novel pricing mechanism inspired by real-world internet architecture that mitigates the severity of Braess' Paradox in selfish routing networks, ensuring Nash equilibria and bounded inefficiency.
Contribution
The paper proposes a new pricing mechanism based on internet graph properties that reduces Braess' Paradox effects in non-cooperative routing games.
Findings
Nash equilibria are proven to exist under the proposed pricing scheme.
Bounds on the inefficiency caused by selfish routing are established.
The mechanism leverages internet's graphical structure to improve network performance.
Abstract
Internet and graphs are very much related. The graphical structure of internet has been studied extensively to provide efficient solutions to routing and other problems. But most of these studies assume a central authority which controls and manages the internet. In the recent years game theoretic models have been proposed which do not require a central authority and the users are assumed to be routing their flows selfishly. The existence of Nash Equilibria, congestion and the amount of inefficiency caused by this selfish routing is a major concern in this field. A type of paradox in the selfish routing networks, Braess' Paradox, first discovered by Braess, is a major contributor to inefficiency. Several pricing mechanisms have also been provided which give a game theoretical model between users(consumers) and ISPs ({Internet Service Providers} or sellers) for the internet. We propose…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsGame Theory and Applications · Merger and Competition Analysis · Consumer Market Behavior and Pricing
