Paper Info
Title
Dynamic Generalization Kanerva Coding in Reinforcement Learning for TCP Congestion Control Design.
Abstract
Traditional reinforcement learning (RL) techniques often encounter limitations when solving large or continuous state-action spaces. Training times needed to explore the very large space are impractically long, and it can be difficult to generalize learned knowledge. A compact representation of the state space is usually generated to solve both problems. However, simple state abstraction often cannot achieve the desired learning quality, while expert state representations usually involve costly hand-crafted strategies. We propose a new technique, generalization-based Kanerva coding, that automatically generates and optimizes state abstractions for learning. When applied to adapting the congestion window of the highly complex TCP congestion control protocol, a standard Internet protocol, this technique outperforms the current standard-TCP New Reno by 59.5% in throughput and 6.5% in delay. Our technique also achieves a 35.2% improvement in throughput over the best previously proposed Kanerva coding technique when applied in the same context.
Year
DOI
Venue
2017
10.5555/3091125.3091375
AAMAS
Keywords
Field
DocType
state abstraction,TCP congestion control,dynamic generalization,Kanerva coding
Internet Protocol,Abstraction,Congestion window,Computer science,Coding (social sciences),Artificial intelligence,Throughput,State space,Machine learning,TCP congestion-avoidance algorithm,Reinforcement learning
Conference
Citations 
PageRank 
References 
0
0.34
7
Authors
4
Name
Order
Citations
PageRank
Wei Li141.11
Fan Zhou2298.68
Waleed Meleis315718.29
Kaushik R. Chowdhury42909144.16