References of "Bommaraveni, Srikanth 50030331"
     in
Bookmark and Share    
Full Text
Peer Reviewed
See detailActive Popularity Learning with Cache Hit Ratio Guarantees using a Matrix Completion Committee
Bommaraveni, Srikanth UL; Vu, Thang Xuan UL; Chatzinotas, Symeon UL et al

Scientific Conference (2020, October 08)

Edge caching is a promising technology to facethe stringent latency requirements and back-haul trafficoverloading in 5G wireless networks. However, acquiringthe contents and modeling the optimal cache ... [more ▼]

Edge caching is a promising technology to facethe stringent latency requirements and back-haul trafficoverloading in 5G wireless networks. However, acquiringthe contents and modeling the optimal cache strategy is achallenging task. In this work, we use an active learningapproach to learn the content popularities since it allowsthe system to leverage the trade-off between explorationand exploitation. Exploration refers to caching new fileswhereas exploitation use known files to cache, to achievea good cache hit ratio. In this paper, we mainly focus tolearn popularities as fast as possible while guaranteeing anoperational cache hit ratio constraint. The effectiveness ofproposed learning and caching policies are demonstratedvia simulation results as a function of variance, cache hitratio and used storage. [less ▲]

Detailed reference viewed: 70 (4 UL)
Full Text
Peer Reviewed
See detailActive Content Popularity Learning via Query-by-Committee for Edge Caching
Bommaraveni, Srikanth UL; Vu, Thang; Vuppala, Satyanarayana et al

Scientific Conference (2019, November 03)

Edge caching has received much attention as an effective solution to face the stringent latency requirements in 5G networks due to the proliferation of handset devices as well as data-hungry applications ... [more ▼]

Edge caching has received much attention as an effective solution to face the stringent latency requirements in 5G networks due to the proliferation of handset devices as well as data-hungry applications. One of the challenges in edge caching systems is to optimally cache strategic contents to maximize the percentage of total requests served by the edge caches. To enable the optimal caching strategy, we propose an Active Learning approach (AL) to learn and design an accurate content request prediction algorithm. Specifically, we use an AL based Query-by-committee (QBC) matrix completion algorithm with a strategy of querying the most informative missing entries of the content popularity matrix. The proposed AL framework leverage's the trade-off between exploration and exploitation of the network, and learn the user's preferences by posing queries or recommendations. Later, it exploits the known information to maximize the system performance. The effectiveness of proposed AL based QBC content learning algorithm is demonstrated via numerical results. [less ▲]

Detailed reference viewed: 72 (12 UL)