Please join the Slack channel.

We are in room 6 on Gather.Town.

Please note that the below times are based on Pacific Time (PT); please convert them to your local time

August 23, 2021 (7 AM – 2 PM PT)

Session 1, Opening – Keynote Talk

Time: 7:00 AM – 8:10 AM PT

Chair: Yunus Donmez, Ericsson Research, Turkey

07:00 Opening remarks by the chairs          

07:10 Keynote Talk: 3GPP Standardization of AI-based Network Automation, David M. Gutierrez Estevez, Samsung

Abstract: This talk will provide a technical overview of the 3GPP standardization efforts on Artificial Intelligence (AI) based network automation, with a special focus on the specification of the Network Data Analytics Function (NWDAF) located within the 5G Core (5GC), whose aim is to provide an end-to-end AI and data analytics framework. The talk will cover the state of the art, namely the standardization agreements achieved during both Release 16 and Release 17, as well as the outlook for the future of Network AI in Release 18 and beyond. Research opportunities and challenges leveraging this framework will also be highlighted during the talk.

Biography: Dr. David M. Gutierrez Estevez is a Chief Engineer at Samsung Electronics R&D Institute UK. David obtained his Engineering Degree in Telecommunications (Hons.) from the Universidad de Granada, Spain, and his M.S. (2011) and Ph.D. (2014) degrees on wireless networks from the Georgia Institute of Technology in Atlanta, GA, USA, under the supervision of Prof. Ian F. Akyildiz. From September 2014 to September 2015, David worked for Huawei Technologies in Silicon Valley on the topic of cloud computing. Previously, David had been with several Fraunhofer Institutes in Germany and with Qualcomm in California for periods ranging from three to fifteen months. In 2016, David joined Samsung and led several large-scale research projects within the 5GPPP ecosystem on 5G RAN, 5G architecture and AI. Since 2019, David is a member of the Samsung global 3GPP delegation in SA2 leading the work on data analytics for network automation while involved in other network architecture aspects.

07:55 Keynote Talk Q&A

Break 10 minutes

Session 2, AI in Network Management

Time:8:20 AM – 9:30 AM PT

Chair:Yunus Dönmez, Ericsson Research, Turkey

08:20 Opening by the session chair        

08:24 Invited Paper: AI-driven Closed-loop Automation in 5G and beyond Mobile Networks (Invited paper) Raouf Boutaba (University of Waterloo); Nashid Shahriar (University of Regina); Mohammad Ali Salahuddin, Shihabur Chowdhury, Niloy Saha, Alexander James (University of Waterloo)

08:46 Practical Automation for Management Planes of Service Provider Infrastructure, Bingzhe Liu (University of Illinois at Urbana-Champaign); Kuan-Yen Chou (UIUC); Pramod Jamkhedkar, Bilal Anwer, Rakesh Sinha (AT&T); Kostas Oikonomou (unaffiliated); Matthew Caesar (UIUC); Brighten Godfrey (UIUC and VMware)

09:08 FedRAN:FederatedMobileEdgeComputingwithDifferentialPrivacy, Aashish Gottipati, Alex Stewart, Jiawen Song, Qianlang Chen (University of Utah)

Break 20 minutes

Session 3, AI for QoS

Time:9:50 AM – 11:00 AM PT

Chair:Ilker Demirkol, Univ.  Politecnica de Catalunya, Spain

09:50 Opening by the session chair        

09:54 Recommending Changes on QoE Factors with Conditional Variational AutoEncoder, Selim Ickin

10:16   Reinforcement Learning and Energy-Aware Routing, Erol Gelenbe, Piotr Frohlich, Mateusz Nowak (Institute Of Theoretical And Applied Informatics Of The Polish Academy Of Sciences)

10:38   Mitigation of Scheduling Violations in Time-Sensitive Networking using Deep Deterministic Policy Gradient, Boyang Zhou (Lehigh University); Liang Cheng (University of Toledo)

Break 20 minutes

Session 4, Cadence Talk

Time:11:20 AM – 00:30 PM PT

Chair:Ilker Demirkol, Univ.  Politecnica de Catalunya, Spain

11:20 Opening by the session chair         

11:30 Cadence Talk:  Drivers and directions for 6G network architecture, Gunnar Mildh, Ericsson Research

Abstract: Research on future network architecture targeting 6G/2030 time frame has started in industry and academia. The presentation will include some high level drivers and directions for future 6G network architecture covering areas such as flexibility, efficient cloud deployment, resiliency and integration of AI/ML.

Biography: Gunnar Mildh received his M. Sc. in electrical engineering from the Royal Institute of Technology (KTH), Stockholm, Sweden, in 2000. In the same year he joined Ericsson Research, Ericsson AB, Stockholm, and has since been working on standardization and concept development for GSM/EDGE, HSPA, LTE, 5G NR and 6G. His focus areas are radio network architecture and protocols. He is currently a senior expert in radio network architecture at the Research Area Networks, Ericsson Research

12:15   Cadence Talk Q&A

Break 20 minutes 

Session 5, AI in Networks

Time: 00:50 PM – 2:00 PM PT

Chair: Ertan Onur, Middle East Technical University, Turkey

12:50 Opening by the session chair       

12:54 Internet Traffic Classification Using an Ensemble of Deep Convolutional Neural Networks, Amin Shahraki (Østfold University College); Mahmoud Abbasi (Islamic Azad Unviersity); Amir Taherkordi (University of Oslo); Mohammed Kaosar (Morduch University)

13:16 Automated Collaborator Selection for Federated Learning with Multi-armed Bandit Agents, Hannes Larsson, Hassam Riaz, Selim Ickin (Ericsson AB)

13:38 A Reinforcement Learning Framework for Optimizing Throughput in DOCSIS Networks, Kevin Dugan, Maher Harb, Daniel Rice (Comcast Cable)

14:00   Closing remarks by the chairs