G4360: Theoretical Neuroscience

• Lecturers: Larry Abbott, Ken Miller, Stefano Fusi


• Tentative topics:
This course will provide a comprehensive introduction to the field of computational neuroscience, covering three broad areas: mathematical techniques, computer simulation methods, and modeling neural systems. Mathematical techniques will include solutions of ordinary differential equations, matrix diagonalization including eigenvalues and eigenvectors, principal components analysis, and information theory. Computer simulations will be done using Matlab and will involve methods of numerical integration of nonlinear differential equations, minimization and stochastic approaches, among other numerical methods. Course exercises will include construction of models of channels, single neurons, synapses, small circuits of neurons and large networks, and explorations of neural coding, selectivity, memory, and supervised, unsupervised and reward-based forms of learning.


• Text: Theoretical Neuroscience By P. Dayan and L.F. Abbott (MIT Press)
  


• Lectures: Wednesdays 2:00-3:30 pm and Fridays 2:30-4:00, starting January 18 in the 7th floor conference room, Kolb Research Annex
  


• Problem Sets
• Problem Set #9 is due on May 2.
• Problem Set #8 is due on April 18.
• Problem Set #7 is due on April 11.
• Problem Set #6 is due on March 28.
• Problem Set #5 is due on March 14.
• Problem Set #4 is due on March 7.
• Problem Set #3 is due on February 14.
• Problem Set #2 is due on February 7.
• Problem Set #1 is due on January 31.


• TA's: OFFICE HOURS
  
• Brian DePasquale
• Saul Kato
• Ann Kennedy
• David Pfau
• Dan Rubin


• Course Syllabus


• Announcements
• 4/11/2011: Class on Wednesday, April 20 will be held in the Hammer 5th floor conference room


• Useful Links:
• Ken Miller's notes on linear algebra. Written for neuroscience students without much math background.
  
• Ken's lecture on information theory and coding.
  
• Ken's lecture notes on decoding.
  
• Ken's lecture notes on optimization.