Vernade Lab

We study the mathematical foundations of interactive and adaptive learning systems. Our work spans reinforcement learning, bandit algorithms, continual learning, and learning theory.

Most learning theory assumes a fixed world and a single objective. We study what happens when the world changes, feedback arrives late or indirectly, and a learner must acquire useful knowledge over a lifetime.

The group is led by Claire Vernade and is part of the University of Technology Nuremberg.

Research areas

• Continual and lifelong learning: agents that reuse experience, adapt to change, and plan what to learn next.
• Reinforcement learning theory: principled algorithms for decision-making under uncertainty.
• Bandit algorithms: efficient exploration with structured, delayed, or partial feedback.

Read more about our research or meet the team.