目前書報討論已改為線上課程，欲旁聽的「電機系」同學可以寄信給助教(happy990044@gmail.com)索取直播連結

Date&Time:

2022 / 05 / 13 (Fri) 14:20 ~ 16:20

Location :

Online Course

Speaker: 

Dr. Amir Rezapour

國立清華大學 資訊安全研究所 (Inst. of Information Security, NTHU)

Topic: 

Towards Optimal Routing via Reinforcement Learning

Abstract:

 Communication networks are difficult to model and predict because they have become very sophisticated and dynamic. We develop a reinforcement learning routing algorithm to solve a traffic engineering problem of SDN in terms of throughput and delay. Our method solves the traffic engineering problem via experience, instead of building an accurate mathematical model. The experimental results show that our method obtains higher rewards and enables a host to transfer a large file faster than Open Shortest Path First (OSPF) and Least Loaded (LL) routing algorithms on various network topologies. For example, on the NSFNet topology, the sum of rewards obtained by our method is 119.30, whereas those of OSPF and LL are 106.59 and 74.76, respectively. The average transmission time for a 40GB file using our method is 25.2s. Those of OSPF and LL are 63s and 53.4s, respectively.

Short Bio:

Amir Rezapour received an M.S. degree in Computer Science from National Tsing Hua University, Taiwan, 2013 and a Ph.D. degree in Computer Science from National Yang Ming Chiao Tung University, Taiwan, 2018. He is currently an Adjunct Assistant Professor at National Tsing Hua University as well as a Security AIML Engineer at Logitech. His research interests are in the area of Cryptography and Network Security.