Philosophy and Artificial Intelligence
Aaron Sloman
The slides used for my lecture on Philosophy and AI can be found here
as promised:
http://www.cs.bham.ac.uk/research/cogaff/talks/#aiandphil
They are available in several formats including a PDF file.
(The slides were written using Latex).
There are many more slides than I presented in the lecture, some of
which may be of interest to students wishing to find out more about
philosophy and how it can interact with AI.
My
talks
website has many other slide presentations that you are welcome to use.
The Birmingham
Cognition and Affect
website has many papers and PhD theses on it.
There is a robotics project called 'CoSy', funded by the Euorpean
Community which involves 7 Universities in different places, and which I
hope will address some of the philosophical issues mentioned in my talk.
The Birmingham website for CoSy is
http://www.cs.bham.ac.uk/research/projects/cosy/
Take a look at this partial specification of the tasks in designing the
robot called
'PlayMate'.
The reactive/deliberative sheepdog demo shown at the beginning of the
lecture was implemented using the SimAgent toolkit, described in:
http://www.cs.bham.ac.uk/research/poplog/packages/simagent.html
This toolkit is written in Pop11 and makes use of several extensions to
Pop11, including ObjectClass (which provides object-oriented programming
using multiple-inheritance), RCLIB (which provides 2-D graphical
interface tools, POPRULEBASE (which provides a sophicated package for
multi-threaded rule-based programming), and the SimAgent library which
puts all the pieces together. The toolkit has been used in many
undergraduate projects. It is included as part of Poplog (on the school
DVD) and is available on our linux PCs and Suns. There are some
demo movies (non-interactive) here:
http://www.cs.bham.ac.uk/research/poplog/figs/simagent/
If you use the toolkit or Pop11 to develop something you would like to
make available for future students to play with and learn from, let me
know and we can try to arrange details.
Did anyone manage to work out the proof that there are infinitely many
prime numbers. The clue I gave you in the lecture was to assume that
there is a largest prime number P, say, and use that assumption to
derive a contradiction by showing how to create a larger one.
It will help if you know what the factorial of a number is.
It will also help if you remember that given any number N there is a
mechanical procedure for finding the smallest number (bigger than 1)
that divides N, and that number must be prime. (Why?)
NOTE:
Please remember: if you sit passively listening in lectures you will not
remember anything later: you need to be active in some way, e.g. taking
rough notes on the content of the lecture, and then later that day or
the next day, take your notes and re-write them in a more coherent form.
After that you can throw them away: the point is not to have the notes
in the end, but to engage with the notes by doing something active.
That's how you will learn -- and stretch your minds, which is presumably
what you came to University to do.
If you think I should provide any more information on this web page,
please let me know.
Aaron (A.Sloman@cs.bham.ac.uk)
Last updated: 27 Oct 2005