Ecologically Inspired Distributed Search in Unstructured Peer-to-Peer Networks

Li Sa, Yongsheng Ding
2007 2007 IFIP International Conference on Network and Parallel Computing Workshops (NPC 2007)  
In this paper, we reported an ecosystem inspired algorithm for searching peer-to-peer networks. An agent-based model had been developed to solve the problems caused by the blind flooding-based search, such as inefficient search and traffic, etc. Reinforcement learning and evolution of the agents can change the agents population toward sampling areas of the environment that are close to resources by increasing the density of agents near those sources.
doi:10.1109/npc.2007.123 fatcat:5hwunpdjbnfdjm7jeuovsvbqgm