EPFL · Adaptive Intelligence

Adaptive intelligence, in brains, bodies and machines.

We study how brains and bodies adapt, and build AI that is robust to the unexpected and interpretable by design.

Research

Four arms, one goal.

Computer vision measures behavior, biomechanics models the body that produces it, neural dynamics asks how the brain controls the body and adapts, and machine learning gives us the methods. Robust, interpretable AI runs through all four themes.

A CEBRA latent embedding of neural data
Neural dynamics

Uncover how the brain controls the body and adapts

How brains build and update internal models during skilled motor learning, studied with large-scale recordings and consistent latent embeddings of neural and behavioral data.

  • CEBRA
  • Internal models
  • Adaptive motor control across species
Machine learning

Make representations consistent, robust and causal

Identifiable representations, governing equations recovered from latent dynamics, attribution you can trust, and models that hold up under adversarial and out-of-distribution inputs.

Robust & interpretable AI Robust to the unexpected · Interpretable by design
How we get there · foundations of AI

Real-world behavior is the ultimate stress test for AI. Scarce labels, long-tailed data and constant distribution shift are not edge cases there; they are the norm. Those constraints drive every method we build.

01

Scarce labels

Foundation models and self-supervised learning that measure behavior across species without exhaustive annotation.

02

Long-tailed data

Multimodal models that understand rare actions and complex scenes, not just the common cases.

03

Distribution shift

Out-of-distribution robustness and identifiable representations that stay consistent when the world changes.

Highlights

Research in motion.

NeurIPS · 2026

DYSCO

The equations of motor cortex, learned from spikes.

DYSCO uses multi-view contrastive learning to recover hidden neural trajectories and the dynamics that govern them, then writes those dynamics down as equations you can read.

Paolo Muratore & Mackenzie Weygandt Mathis

Latest publications

Recent papers.

Full list
    News

    What's new in the lab.

    Full timeline
      Join the M-Lab

      Build AI that holds up in the real world, and/or study how real brains adapt.

      We are looking for PhD students, postdocs, engineers and students who want to work at the intersection of machine learning, computer vision, neuroscience and biomechanics.

      Funding

      With thanks to our supporters.

      Our work is made possible by current and past support from:

      Bertarelli Foundation
      Swiss National Science Foundation
      Chan Zuckerberg Initiative
      The Vallee Foundation
      National Institutes of Health
      Google AI
      NVIDIA
      Harvard University
      • Swiss National Science Foundation (SNSF)
      • Bertarelli Foundation
      • Simons Foundation
      • Kavli Foundation
      • Chan Zuckerberg Initiative
      • Wellcome Trust
      • The Vallee Foundation
      • NIH BRAIN Initiative
      • Radala Foundation
      • Google AI
      • Novartis Foundation for Medical-Biological Research
      • Wyss Center
      • NVIDIA
      • Harvard Brain Science Initiative
      • Mind, Brain, Behavior Interfaculty Initiative
      • Rowland Institute at Harvard
      • Women & the Brain / Project ALS
      • National Science Foundation (GRFP)
      • The Imaging Source, LLC