Daniel Levenstein, PhD

Like learning, sleep changes the brain to improve its future performance. Unlike learning, these changes occur in the absence of overt behavior or sensory input. This "offline learning" thus contains a mystery: how does internally-generated activity improve brain function? This question lies at the intersection of biology, dynamics, and computation – it requires connecting the emergent organization of neural activity during sleep with the operations it performs on the brain's information processing capacities. Daniel Levenstein's lab aims to tackle this problem by building artificial intelligence systems that mimic spontaneous activity in the sleeping brain and its use for offline learning. In addition to their work on sleep, the lab works broadly at the interface of theoretical and experimental neuroscience - applying computational methods to a variety of interesting problems involving neural dynamics and computation with experimentalist collaborators.

biophysics ;  computational modeling ;  deep learning ;  neural networks ;  simulation  

artificial intelligence ;  computational neuroscience ;  long-term memory ;  machine learning ;  navigation ;  philosophy ;  sleep ;  systems neuroscience  

Levenstein received his Bachelor's degree in Biochemistry with a minor in Physics from Northeastern University, his Master's in Biophysics from Cornell University, and his PhD in Neuroscience from New York University under the mentorship of György Buzsáki and John Rinzel. He completed his postdoctoral training at the interface of neuroscience and artificial intelligence at McGill University and Mila (the Quebec AI Institute), working with Adrien Peyrache and Blake Richards. Levenstein started his lab at Yale in 2025, in the Department of Neuroscience. When he's not in the lab, he enjoys hiking, board games, weird music, and a good book.