3rd ACM Workshop on Machine Learning for Next Generation Communication and Edge Networks (ML4NxtGNet)

Co-located with ACM MobiHoc 2026

Tokyo, Japan, November 26, 2026

• 8:25 am - Welcome & Overview of Workshop 

• 8:30 am - Invited Talk 1 - Igor Kadota, Northwestern University

Fair Dynamic Spectrum Access via Fully Decentralized Multi-Agent Reinforcement Learning 

Abstract: This talk presents Fair Share Reinforcement Learning (FSRL), a fully decentralized multi-agent reinforcement learning approach to dynamic spectrum access. Each agent independently selects when and where to transmit using only the outcomes of its own previous actions, without knowing the number of competing agents or exchanging information explicitly or implicitly. To promote fair spectrum sharing under these stringent information constraints, FSRL combines an augmented state with a semi-adaptive time reference, a distributional reinforcement learning architecture incorporating dynamic risk and time-difference likelihood, and a fairness-driven reward structure. We evaluate FSRL across 54 network configurations with different numbers of agents and frequency bands, as well as under jamming and ad hoc interference. The results show that decentralized agents can learn efficient and equitable transmission patterns without centralized training or coordination. 

Bio: Igor Kadota is an Assistant Professor of ECE at Northwestern University. He received the Ph.D. degree from MIT LIDS and his B.Sc. degree from ITA in Brazil. His research is on modeling, analysis, optimization, and implementation of next-generation communication networks, with the emphasis on advanced wireless systems and time-sensitive applications. He was a recipient of research, teaching, and mentoring awards, including the Best Paper Award at IEEE INFOCOM 2018, and the Best Student Paper Award at WiOpt 2024 and at WiOpt 2025.

• 9:00 am - Invited Talk 2 - Stratis Ioannidis, Northeastern University

• 9:30 am - Invited Talk 3 - Takayuki Nishio, Institute of Science Tokyo

• 10:00 am - Accepted Paper

• 10:20 am - Coffee Break

• 10:40 am - Invited Talk 4 - Jian Li, Stony Brook University

• 11:10 am - Invited Talk 5 - Brian M. Kurkoski, Japan Advanced Institute of Science and Technology

• 11:40 am - Invited Talk 6 - Suguru Kameda, Hiroshima University

Synchronization-Oriented Multiple Access (SOMA): Leveraging Space-Time Synchronization for Massive Connect IoT

Abstract: Future wireless networks must efficiently support massive numbers of devices and diverse data types. This talk introduces Synchronization-Oriented Multiple Access (SOMA), which leverages high-precision space-time synchronization to structure interference and improve communication efficiency, thereby addressing the scalability limitations of conventional multiple access schemes that treat asynchronism as an impairment. The talk reviews space-time synchronization using Wireless Two-Way Interferometry (Wi-Wi) modules provided by the National Institute of Information and Communications Technology (NICT) and presents synchronized spread-spectrum code-division multiple access (SS-CDMA) as one realization of SOMA for Massive Connect Internet of Things (Massive Connect IoT) scenarios. SOMA provides a scalable multiple-access framework for next-generation dense IoT networks. 

Bio: Suguru Kameda received the B.S., M.S., and Ph.D. degrees in electronics engineering from Tohoku University, Sendai, Japan, in 1997, 1999, and 2001, respectively. In 2001, he joined the Research Institute of Electrical Communication (RIEC), Tohoku University. Since 2021, he has been a full professor at Hiroshima University, Higashi-Hiroshima, Japan, and is currently affiliated with the Research Institute for Semiconductor Engineering (RISE). From 2018 to 2019, he was a visiting scholar at the Wireless Information Network Laboratory (WINLAB), Rutgers University, NJ, USA. From 2018 to 2021, he was an invited advisor at the National Institute of Information and Communications Technology (NICT), Tokyo, Japan. His current research interests include massive connect IoT, heterogeneous networks, and satellite communications. He is a Fellow of IEICE, a Senior Member of IEEE, and a member of JIEP. He serves as Chair of the IEEE Hiroshima Section for 2025-2026.

• 12:10pm - Closing Workshop Remarks & Workshop Adjourns