9.641 Neural Networks

To view the equations in the HTML-version of the lectures with a UNIX-browser you will have to configure X according to these instructions
Get the latest free version of RealPlayer here

The lectures are all taped, and are available in RealVideo format. Each lecture is linked below, but if you'd like to download them, they're here.

Lecture schedule

• Lecture 1 (2/1): From spikes to rates (Postscript, PDF, realvideo)
  • Christof Koch, Biophysics of Computation, Section 14.2: 335-341, Oxford University Press, New York, Oxford (1999).
  • Bard Ermentrout, Reduction of Conductance-Based Models with Slow Synapses to Neural Nets, Neural Computation. 6,679-695 (1995).
• Lecture 2 (2/3): Perceptrons: simple and multilayer (realvideo)
• Lecture 3 (2/8): Perceptrons as models of vision (realvideo)
  
  • D. Marr, Vision, Section 2.2, 54-79, W.H. Freeman and Company, New York (1982).
  • D.H. Hubel, Eye, Brain, and Vision, Chapter 3, 39-46, Scientific American Library, New York, (1988-1995).
  • Lenet website
• Lecture 4 (2/10): Linear networks (realvideo)
• Lecture 5 (2/15): Retina (realvideo)
  • E.H. Adelson, Lightness Perception and Lightness Illusions, In The New Cognitive Neurosciences, 2nd ed., M. Gazzaniga, ed. Cambridge, MA: MIT Press, pp. 339-351, (2000).
  • H. K. Hartline and F. Ratliff, Inhibitory Interaction in the Retina of Limulus. Physiology of Photoreceptor Organs, Ed. M.G.F. Fuortes, 382-447, Springer-Verlag, Berlin, Heidelberg, New York (1972)
• Lecture 6 (2/17): Lateral inhibition and feature selectivity (slides 1, slides 2, slides 3, realvideo)
  • W. H. Press et al., Numerical Recipes in C, 2d ed., Chaps. 12 and 13.
  • G. Strang, Introduction to Applied Mathematics, Section 4.2, 290-309, Wellesley-Cambridge Press, Wellesley, Massachusetts (1986).
• Lecture 7 (2/24): Objectives and optimization (realvideo)
• Lecture 8 (3/1): Hybrid analog-digital computation, ring network (realvideo)
  • R. Hahnloser, R. Sarpeshkar, M. Mahowald, R.J. Douglas, H.S. Seung, Digital selection and analog amplification coexist in a cortex-inspired silicon circuit, Nature, 405, 947-51, 2000.
  • Richard H. R. Hahnloser, H. Sebastian Seung, Jean-Jacques Slotine: Permitted and Forbidden Sets in Symmetric Threshold-Linear Networks. Neural Computation 2003.
• Lecture 9 (3/3): Constraint satisfaction, stereopsis (realvideo)
• Lecture 10 (3/8): Bidirectional perception (realvideo)
• Lecture 11 (3/10): Signal reconstruction (realvideo)
• Lecture 12 (3/15): Hamiltonian dynamics (slides, realvideo)


• Midterm (3/17)


• Lecture 13 (3/29): Antisymmetric networks (slides, realvideo)
• Lecture 14 (3/31): Excitatory-Inhibitory networks, learning (slides, realvideo)
• Lecture 15 (4/5): Associative memory (realvideo)
• Lecture 16 (4/7): Models of delay activity, Integrators (realvideo)
• Lecture 17 (4/12): Multistability, Clustering (realvideo)
• Lecture 18 (4/14): VQ and PCA (slides 1, slides 2, realvideo)
• Lecture 19 (4/21): More PCA, Delta rule (slides, realvideo)
• Lecture 20 (4/26): Conditioning and Backpropagation (slides 1, slides 2, realvideo)
• Lecture 21 (4/28): More Backpropagation (slides, realvideo)
• Lecture 22 (5/3): Stochastic Gradient Descent (realvideo)
• Lecture 23 (5/5): Reinforcement Learning (realvideo)
• Lecture 24 (5/10): More Reinforcement Learning (realvideo)
• Lecture 25 (5/12): Final Review (realvideo)