Forgis Research Night – Learning to Dream in EEG

Code Slides Slides (LuMamba / LeJEPA) BioFoundation codebase LUNA paper (arXiv)
Learning to Dream in EEG — Forgis Research Night, DARE Campus

Abstract

Invited lightning talk at Forgis Research Night (DARE Campus, Schlieren). I introduced LuMamba: a 4.6M-parameter EEG foundation model that pairs LUNA’s channel-unification module — learned queries that read any electrode montage into a fixed latent space — with a linear-time bidirectional Mamba backbone. The core idea is to “learn to dream”: instead of reconstructing the raw, noisy signal, we borrow the representation-learning half of LeCun’s world-model recipe and predict in latent space with LeJEPA, kept collapse-free by the SIGReg (Sketched Isotropic Gaussian) regularizer. Mixing a little masked reconstruction with LeJEPA keeps clusters and generalizes, reaching state-of-the-art Alzheimer’s detection (0.97 AUPR) — strongest on montages never seen during pretraining.

Date
24, Jun, 2026 64:00
Event
Forgis Research Night
Location
DARE Campus (The JED), Zürich-Schlieren, Switzerland
The JED, Schlieren, Canton of Zürich 8952

A short lightning talk on how we pretrain foundation models for biosignals — and why it pays to teach the model to dream rather than to copy.

What I covered

  • The problem: EEG is a tiny (~20 µV) signal buried in noise (SNR < 1), and every dataset uses a different electrode montage.
  • LuMamba: LUNA channel-unification (learned queries → fixed latent, reads any montage) + a linear-time bidirectional Mamba backbone — 4.6M parameters.
  • Learning to dream: rather than reconstructing the raw signal, we predict in latent space with LeJEPA, kept collapse-free by SIGReg. A mixed reconstruction + LeJEPA objective keeps cluster structure and generalizes — state-of-the-art Alzheimer’s detection (0.97 AUPR), strongest on unseen montages.

Slides & resources

Photos

Kicking off — *Learning to Dream in EEG* at DARE Campus.
Kicking off — Learning to Dream in EEG at DARE Campus.
A full room for Forgis Research Night.
A full room for Forgis Research Night.
Forgis Foundation Models EEG Self-Supervised Learning LeJEPA World Models
Thorir Mar Ingolfsson
Thorir Mar Ingolfsson
Postdoctoral Researcher

I develop efficient machine learning systems for biomedical wearables that operate under extreme resource constraints. My work bridges foundation models, neural architecture design, and edge deployment to enable real-time biosignal analysis on microwatt-scale devices.

Related