Applied Learning Networks (ALN)
Abstract
Applied Learning Networks (ALN) demonstrates that a network protocol can learn to improve its performance over time, showing how to incorporate learning methods into a general class of network protocols. ALN applies accumulated experience with previous network connections to help tune future network connections. ALN provides demonstration of non-trivial learning in complex communication protocols. It also provides proof that learning results in task specific performance enhancements.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 2007
- Accession Number
- ADA462328
Entities
People
- Feili Hou
- Joseph Bannister
- Joseph Touch
- Venkata Pingali
- Wei-min Shen
Organizations
- University of Southern California