Increasing PageRank through Reinforcement (2002) [2 citations — 1 self]
Popular Tags
No tags have been applied to this document.
BibTeX | Add To MetaCart
author = {Learning Adrian Agogino and Adrian K. Agogino and Joydeep Ghosh},
title = {Increasing PageRank through Reinforcement},
booktitle = {In Proceedings of Intelligent Engineering Systems Through Artificial Neural Networks (St},
year = {2002},
pages = {27--32},
publisher = {ASME Press}
}
Bookmarks
OpenURL
Abstract:
This paper describes a reinforcement learning method, derived from collective intelligence principles, for increasing the combined PageRank for a set of domains. This increased rank is achieved through a set of cooperating reinforcement learners that learn, through exploration, how to add links within the set of domains. We show how reinforcement learners using traditional reward functions perform very poorly at this task. However, reinforcement learners that use rewards based on collective intelligence can achieve good results. The reinforcement learners have an advantage over standard optimization methods in that they can learn with highly non-linear constraints. Additionally we demonstrate that reinforcement learners are robust in that high values of PageRank can be reached, even if all the domains are not cooperating properly.
Citations
| 1064 | The PageRank Citation Ranking: Bringing Order to the Web – Page, Brin, et al. - 1999 |
| 82 | Stable algorithms for link analysis – Ng, Zheng, et al. - 2001 |
| 59 | An Introduction to Collective Intelligence – Wolpert, Tumer - 1999 |
| 46 | Optimal payoff functions for members of collectives – Wolpert, Tumer |
