Accession Number:

AD1017108

Title:

Stochastic Online Learning in Dynamic Networks under Unknown Models

Descriptive Note:

[Technical Report, Final Report]

Corporate Author:

University of California - Davis

Personal Author(s):

• Zhao, Qing

Report Date:

2016-08-02

Pagination or Media Count:

15

Abstract:

This research aims to develop fundamental theories and practical algorithms for distributed, robust, and real-time learning in dynamic tactical networks. The overall objective is to significantly move the frontiers of knowledge in stochastic learning in the classic multi-armed bandit by systematically relaxing traditionally adopted restrictive assumptions.

Descriptors:

• mesh networks
• signal processing
• information theory
• stochastic processes
• algorithms
• random variables
• ad hoc networks
• communication networks
• communication systems
• engineering
• information exchange
• information processing
• network topology
• networks
• computers
• department of defense
• dimensionality reduction
• mathematics
• probability
• spine

Subject Categories:

• Operations Research
• Cybernetics

Distribution Statement:

[A, Approved For Public Release]