BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Service-Aware Provisioning for Greener &amp\; Smarter Cellular Net
 works - Li Rongpeng\, Zhejiang University\, Hangzhou China
DTSTART:20200625T140000Z
DTEND:20200625T150000Z
UID:TALK142783@talks.cam.ac.uk
CONTACT:Srinivasan Keshav
DESCRIPTION:Cellular Networks has evolved into the fundamental infrastruct
 ure of the information and communication technology (ICT) industry. Faced 
 with the increasingly diversified service demands and heterogeneous networ
 k architecture\, it is becoming essential to further integrate the artific
 ial intelligence and cellular networks. In this talk\, I will first provid
 e a general overview of our research on service-aware provisioning\, which
  we believe is an essential ingredient for intelligent cellular networks. 
 Afterwards\, I will dip into one of my recent works on reinforcement learn
 ing-based resource management for network slicing. Specifically\, we consi
 der a scenario that contains several slices in a radio access network with
  base stations that share the same physical resources (e.g.\, bandwidth or
  slots)\, and try to dynamically and efficiently allocate resources for di
 versified services with distinct requirements over a common underlying phy
 sical infrastructure. Therein\, we leverage deep reinforcement learning to
  solve this problem by considering the varying service demands as the envi
 ronment state and the allocated resources as the environment action. In or
 der to reduce the effects of the annoying randomness and noise embedded in
  the received service level agreement satisfaction ratio and spectrum effi
 ciency\, we primarily propose generative adversarial network-powered deep 
 distributional Q network (GAN-DDQN) to learn the action-value distribution
  driven by minimizing the discrepancy between the estimated action-value d
 istribution and the target action-value distribution. Finally\, we verify 
 the performance of the proposed GAN-DDQN algorithms through extensive simu
 lations.\n\nBio: Dr. Li is now an assistant professor in College of Inform
 ation Science and Electronic Engineering\, Zhejiang University\, Hangzhou 
 China. From August 2015 to September 2016\, he was a research engineer wit
 h Wireless Communication Laboratory\, Huawei Technologies Co. Ltd.\, Shang
 hai\, China. His research interests currently focus on Reinforcement Learn
 ing\, Data Mining and all broad-sense network problems (e.g.\, resource ma
 nagement\, security\, etc) and he has authored/coauthored tens of papers i
 n the related fields. He serves as an Editor of China Communications.\n
LOCATION:https://meetingsemea10.webex.com/meet/sk818 
END:VEVENT
END:VCALENDAR