Assignment
- Watch “But what is a neural network?” and “Gradient descent: how neural networks learn” and “Backpropagation: Step by Step” (20 mins each).
- Read and annotate this classic commentary by Francis Crick entitled “The recent excitement about neural networks.”
Write out three or more salient quotes from the paper and provide a worthy response (e.g., an astute observation, nuanced critique, or thought-provoking question). Upload your work to Bb as Assignment #3.
Bring a copy of your quotes and responses to class and be prepared to share your thoughts.
Going deeper – optional reading/watching
- As a response to our discussion of the amphibian retina and feature detectors, read Genealogy of the “Grandmother Cell” — a short history of neuroscience essay (30 min).
- As continuation of our discussion of “sparse coding,” watch the seminar “From Natural Scene Statistics to Models of Neural Coding & Representation” by Bruno Olshausen (part 1 only, duration 60 min). This talk was part of a summer school on deep learning and feature detection. The target audience is graduate students studying visual neuroscience and/or computer vision.
Further Watching and Reading
- A video on the mathematics of backpropagation (20 min).
- The first half of this lecture on “Convolutional Neural Networks” is recommended (30 min).
- Another good video on Convolutional Neural Networks:
Teaching deep learning with Matlab
Anderson, J.A. and Rosenfeld, E. eds., 2000. Talking nets: An oral history of neural networks. MiT Press. [Amazon] [SWEM]
Computational Neuroscience 