In this course, we will describe some of the most recent attempts to
build computers whose design and operation are modelled after the
In doing so, we shall attempt to address the question:
To answer this question we shall need to ask ourselves some further
- Will it ever be possible to build a machine which possesses a mind?
To start off, we shall go on a brief tour of the Biology of the Brain.
We will describe the overall structure of the
human brain and the elements of which it is made - nerve cells
or neurons. We will see how neurons are connected up into
complicated networks, whose function is to process electrical signals
which constitute all brain activity.
- Is the Mind a machine?
- What is intelligence?
- Can computers think?
- Is there any notion of personal responsibility
for a computer?
Motivated by this biological description, we will examine artificial
Neural Networks, which have been
proposed as an alternative route to achieving artificial
intelligence. We will try to describe how two such popular networks
work - the Hopfield and Perceptron models. These
networks can perform tasks such as memory recall, pattern recognition,
and simple decision making. Some of the features of these networks,
particularly those associated to learning are very reminiscent of
similar processes in the brain.
Finally, we shall discuss some of the features
we would expect to find in an intelligent machine.
A discussion of these issues in artificial
intelligence will constitute the last part of this course.
- Are consciousness and the mind somehow distinct
from the physical workings of the brain?
- How might we ascertain whether a machine was conscious?
- What must a neural network be capable of to
stand a chance of being truly intelligent?