The First ESIT-D2I Competition
Channel Charting · Robust Learning · Unlearning Noisy Contributions
A student competition at the intersection of data science, machine learning, and information theory — held in conjunction with ESIT 2026, Nordfjordeid, Norway.
Register NowThis year's challenge is inspired by the Viking legend of Bifröst — the shimmering bridge connecting Midgard to Asgard, where Valkyries ride across the sky to bring fallen heroes to Valhalla.
Back then, reaching Asgard required divine intervention, a winged horse, and — ideally — having died heroically in battle. Today's Nordic explorers, however, have 5G coverage reaching roughly 99.7% of households in Norway.
So we ask a simple question: If somewhere between Midgard and Asgard there were at least a stable cellular connection, and you were given noisy, partial, and imperfect measurements of the wireless channel — can you chart your way to Bifröst?
Welcome to the first edition of the European School of Information Theory Data-to-Information (D2I) Competition, organized by the IEEE Information Theory Society Student and Outreach Committee. Participants will work with real-world Channel State Information (CSI) measurements, designing methods that are robust to label noise, feature corruption, and faulty data contributions.
Everything you need is one click away. Join the community on Slack for announcements and team-building, then head to the two Kaggle competitions, where the data, baseline code, submission interface, and live leaderboards are hosted.
Phase 1 · In-Person
Phase 2 · Online
Final Session · Online
Participants must register through the online form by June 3, 2026. During registration, you will be asked to indicate whether you are registering as an individual or as part of a team, and whether you plan to attend ESIT in person.
Team formation will be finalized on June 4, after the in-person ESIT session — providing an opportunity for participants to meet and form teams on-site. Each team must include at least one member attending ESIT in person. Participants who cannot attend ESIT should still register by May 26; organizers will assist in forming teams for remote participants.
Final live event sponsored by ELLIS Society, Madrid Unit.
Participants will use a dataset collected in a realistic wireless environment (the DICHASUS dataset, Arena2036 research campus) where multiple remote antenna arrays capture Channel State Information (CSI) measurements. CSI describes how a wireless signal propagates through the environment — capturing reflections, scattering, and attenuation — and can be seen as a rich "fingerprint" of location.
The dataset is divided into two components:
Full dataset and tutorials: dichasus.inue.uni-stuttgart.de. All data, utilities, and the submission interface are hosted on Kaggle — see the Platforms section above for the Task 1 and Task 2 competition pages.
We gratefully acknowledge Prof. Stephan ten Brink, Florian Euchner, and Phillip Stephan for building and documenting this dataset and making it freely available to the research community.
Task 0 · Foundations (discussed at ESIT)
This preliminary task introduces the core elements of CSI-based localization and serves as a warm-up for the competition. Participants will explore basic localization (mapping CSI to positions via supervised learning), basic channel charting (low-dimensional representations preserving spatial structure), and trajectory-aware processing (temporal consistency across consecutive measurements).
Task 1 · Main Challenge
In realistic scenarios, both labels and measurements are imperfect. Participants design models that remain accurate and robust when data quality degrades. Two types of corruption are considered:
The goal is to balance accuracy and reliability, avoiding catastrophic localization errors while maintaining good average performance.
→ Compete on Kaggle: ESIT-D2I · Task 1 ↗
Task 2 · Advanced Challenge
A subset of users contributes corrupted trajectories due to faulty GPS or sensor drift. Participants will:
→ Compete on Kaggle: ESIT-D2I · Task 2 ↗
High localization accuracy is crucial, but equally critical is minimizing catastrophic failures. The combined loss function balances two metrics:
Average L2 distance between true and predicted positions. Reflects overall accuracy.
Smallest radius enclosing 90% of predicted locations around true positions. Measures reliability and consistency.
Performance is evaluated in terms of the lowest achievable combined loss. Submissions are made through Kaggle.
The competition is open to students and researchers worldwide. Participants may compete individually or as a team. Each team must include at least one student who is a participant at ESIT. Teams are limited to a maximum of five participants.
While participation is open globally, prize eligibility may be subject to legal, financial, or regulatory constraints. Prize payments must comply with applicable international regulations. In cases where payment cannot be completed due to regulatory constraints, alternative arrangements will be considered when feasible.
Presenting is mandatory to win. At least one member of each winning team must present their solution at the live ESIT-D2I 2026 results session (in person or remotely, as arranged with the organizers); a team that does not present forfeits its prize. Winners must also deliver reproducible code and documentation so that the organizers can independently verify the results before a prize is awarded. Full conditions are set out in the official competition rules on Kaggle.
Prizes are awarded per team. Each winning team designates one representative to receive the prize on behalf of the team. Distribution within the team is the responsibility of its members.
Organizing Committee
Technical Support & Competition Design