Themed Session Announcements
In the context of the invited session “Communication-efficient gradient compression and coding in distributed learning” at ITW, Yu-Chih (Jerry) Huang [NYCU], Shih-Chun Lin [NTU], and Stefano Rini [NYCU] will organize a session for student research presentations and also a competition.
Update (November 3, 2022): The following teams are the winners of the competition.
Maximilian Egger
Congratulations to the winners!
The details regarding these events are below.
1. Call for Student Research Presentations:
Students are invited to present their research videos on topics related to the themed session. An extended abstract needs to be submitted (submission link) upon the acceptance of which the research videos can be submitted. Please see the CFP for more details about the submission requisites, associated deadlines, and contact details of organizers.
Eligibility for participation:
The main author and presenter must be students, also only original research results will be accepted. Also, at least one of the presenters of the research videos must be registered for ITW 2022 by the time of submission of the final videos. Please note that the registration deadline for ITW is 21st October, 2022, but the camera-ready (final) videos are due by 19th October.
Important dates:
Extended abstract submission deadline: 2 Sep, 2022 (along with an early draft of the presentation)
Acceptance Notification: 30 Sep, 2022
Camera Ready video submission deadline for accepted abstracts: 19 Oct, 2022 (at least one author should be registered for ITW 2022 before this)
Live 5 minute session: To be announced (during ITW event).
2. Competition: Federated Deep learning for CSI estimation in Massive MIMO environments
Description:
Distributed machine learning methods are poised to drastically improve the performance of many aspects of communication engineering – from the physical layer to the application one – by leveraging the richness in the data collected at the user equipment. In this competition, we focus on the problem of federated training of a deep CSI compressor for massive MIMO in 5G protocols and beyond.
Objective:
A set of remote users observe a set of pilot signals as transmitted by a MIMO base station (BS) and are tasked with the distributed training of a compressor for the channel estimate. The training of this compressor occurs in a distributed manner, with the BS orchestrating the training and maintaining a centralized model. Training must occur within a set communication budget and model size.
Important dates:
- Registration opens Aug 10, 2022
- Registration closes Oct 1, 2022
- Evaluation period begins Oct 11, 2022
- Competition results are announced Oct 22, 2022
Eligibility for participation:
The team can consist of up to three members, of which there should be at least two student members. At least one of the student members of the competing team should be registered as a member of the IEEE Information Theory Society at the time of registering for the competition.
Registration and Data for Training:
Competition Sponsors


