Introduction to Connectionism
Neural networks
Overview
Many neurons have elaborate arborizations
The axon is covered with myelin sheaths for faster conductivity
With single-cell recordings, action potentials (spikes) can be recorded
McCulloch-Pitts (1943) Neuron
Neural networks abstract from the details of real neurons
Artificial `neuron'
How to `program' neural networks?
Neural networks and David Marr's model (1969)
Hebb (1949)
William James (1890)
The Hebb rule is found with long term potentiation (LTP) in the hippocampus
With Hebbian learning, two learning methods are possible
Another type of network is based on error-correcting learning
We will look at an example of each type of Hebbian learning
Example of competitive learning: Stimulus `at' is presented
Example of competitive learning: Competition starts at category level
Example of competitive learning: Competition resolves
Example of competitive learning: Hebbian learning takes place
Presenting `to' leads to activation of category node 1
Category 1 is established through Hebbian learning as well
Willshaw networks
Example of a simple heteroassociative memory of the Willshaw type
Example of pattern retrieval
Example of successful pattern completion using a subpattern
Example graceful degradation: small lesions have small effects
Summing up
Email: jaap@murre.com
Home Page: http://www.neuromod.org/courses/public.html
Other information: neuroMod: Home of the Neural and Cognitive Modeling Group at the University of Amsterdam.
Download presentation source
University of Amsterdam
