Dhruva Karkada

• PhD Student, Physics Department
  

dkarkada@berkeley.edu Personal Site

Current Research

Deep learning is humanity’s most successful attempt thus far to imitate human intelligence. Despite fundamental differences between deep learning systems and biological brains, deep learning remains a theoretically- and experimentally-accessible playground for understanding learning as a general phenomenon. The long-term goal of my research is to probe deep learning systems (using both theoretical tools and numerical experiments) to elucidate general properties of systems that learn.

Even as a toy model of learning, deep learning is itself mysterious in many ways. Experiments reveal many interesting behaviors (e.g. feature learning, neural scaling laws, emergent abilities) which are poorly understood from a theory standpoint. In particular, I’m interested the large-learning-rate phenomena associated with feature learning (e.g., dynamics of the local loss geometry, representation alignment, and edge-of-stability behavior) and I hope to understand why deep learning is more sample efficient than kernel machines.

Bio

I grew up in Texas and studied physics, astronomy, and computer science at UT Austin. I joined Berkeley as a physics PhD student in 2021, excited to use ideas from physics to understand information processing in neural networks. In my free time, I enjoy hanging out with friends, cooking, playing chess, and messing with synthesizers.