|
|
I am a computational scientist broadly interested to reverse engineer how the brain and mind work in health and
disease in order to understand the circuits and patterns of neural activity that give
rise to mental experience and behavior in order to design and develop more efficient
intelligent methods and systems for complex data analysis.
As advances in technologies are producing large, complex data sets at an unprecedented rate,
novel theoretical and analytical approaches are needed to realize the potential of these rich
datasets. Understanding neural circuitry requires an understanding of the algorithms and mechanisms
that govern information processing within a circuit and between interacting circuits in the brain
as a whole. Informed by rich observations, formalized theoretical frameworks allow researchers to
infer general principles of brain function and the algorithms underlying functioning neural circuitry.
Theory coupled with mathematical modeling and simulation approaches are needed to identify gaps in
knowledge, to drive the systematic collection of the future data (e.g. so that the collected data
specifically address the model parameters), and to formulate testable hypotheses of neural circuit
mechanisms and how they govern behavioral and cognitive processes. New data analysis methods are
needed to detect features in complex data, often spanning multiple modalities and scales, to reveal
underlying mechanisms of brain function.
My approach in computational modeling is that a top-down theorist. I identify the problem in its
most abstract form, then derive the algorithm that solves this problem, and finally look at how
the brain implements the algorithm. Any successful computational model should first be constrained
by large amounts of data, before it makes any further theoretical predictions, because otherwise
too many plausible alternatives cannot be ruled out. A theory that hopes to link brain to behavior
thus needs to discover the computational level on which brain dynamics control behavioral success.
Research aims to develop:
- Theories, ideas and conceptual frameworks
- Multiscale models to integrate information across large temporal and spatial scales in the nervous system
- Intelligent new methods for complex data analysis
- Intelligent machines with autonomous and creative behavior
|
|
|
|
| |
|
