Dynamic Content Caching and Delivery using Restless Multi-armed Bandits
Centre for Networked Intelligence, IISc via YouTube
Overview
Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Attend a technical seminar exploring advanced content caching strategies using restless multi-armed bandit algorithms. Learn how to optimize dynamic content caching and delivery systems where contents are updated at central servers and cached locally at base stations. Discover the mathematical formulation of optimal content fetching problems that minimize average costs while managing cache capacity constraints. Explore the continuous time restless multi-armed bandit process (RMAB) framework where each arm represents different content. Understand the concept of indexability in single content problems and examine Whittle index-based solutions for joint optimization challenges. Investigate the "wait" option strategy that allows base stations to queue additional requests before fetching fresh content, and analyze how this approach changes system state dynamics and caching policies. Compare proposed policies against recent research works and examine simulation results demonstrating near-optimal performance. Gain insights into communication networks, edge caching, edge computing, and optimization techniques relevant to modern content delivery systems.
Syllabus
Time: 5:30 PM - 6:30 PM IST
Taught by
Centre for Networked Intelligence, IISc