Michael DeWeese

Michael DeWeese is an Associate Professor of Physics and Neuroscience at the University of California, Berkeley. His research spans nonequilibrium statistical mechanics, machine learning theory, and systems neuroscience. He holds a BA in Physics from UC Santa Cruz (1988) and a PhD in Physics from Princeton (1995). Before joining UC Berkeley in 2007, he held postdoctoral positions at the Salk Institute and Cold Spring Harbor Laboratory. His work integrates principles from physics, neuroscience, and machine learning to address fundamental questions in theoretical and experimental biology, computation, and statistical mechanics. DeWeese Lab Website provides further details on ongoing projects and collaborations.

Education:

• BA in Physics, UC Santa Cruz (1988)
• PhD in Physics, Princeton University (1995)

Research Interests:

• Nonequilibrium Statistical Mechanics: Focuses on thermodynamic optimization, active matter, and non-equilibrium processes.
• Machine Learning Theory: Develops first-principles models to explain neural network performance and efficient algorithms for probabilistic models.
• Systems Neuroscience: Uses biologically inspired models to understand neural coding, sensory processing, and computational roles of neural networks.

Advising & Grants: While no formal student advisees are listed, his lab actively collaborates across disciplines. Funding sources are not explicitly mentioned but likely involve NSF, NIH, or DOE grants based on research themes. His work on quantum control and neural networks suggests potential ties to interdisciplinary funding initiatives.

Labs & Teams: Directs the DeWeese Lab, which bridges physics, neuroscience, and machine learning. Collaborations include institutions like the Helen Wills Neuroscience Institute (UC Berkeley).