I study minds as computational systems—entities that adapt and act purposefully in uncertain, non-stationary, open-ended environments.
I’m interested in a broad range of problems such as continual learning, hierarchical planning, survival-critical adaptive behaviour and homeostasis. I'm curious about questions such as how the brain does so much with so little! And why do we sometimes cope well with adverse and stressful events, while sometimes we don't?
I draw inspiration from AI research to (1) better understand cognition from a normative perspective and to (2) view mental health disorders and chronic pain as disruptions in biological continual learning, opening paths for new treatments. On the flip side, by drawing on neuroscience, I aim to (3) advance fundamental AI research and (4) develop tools to mechanistically interpret complex AI systems and develop theory-driven, trustworthy AI systems for digital health.
My research has the following research themes -
Computational (Cognitive) Neuroscience of adaptive behaviour (e.g., reinforcement learning in non-stationary environments)
Computational (Cognitive) Neuroscience of homeostasis (e.g., theoretical models of what unfolds after you get injured)
Brain-inspired AI and machine learning (i.e, how can neuroscience help solve the unsolved problems in machine learning?)
Cognitive science-inspired evaluation, interpretability, and development of frontier AI
Theory-driven and human-centred AI for digital health (e.g., digital twins for longitudinal prediction)
Use of emerging technologies in experimental neuroscience (e.g., virtual reality, robotics-based test beds)