Organisers: Owen Holland and
Anil Seth
Tutors:
Owen Holland and Anil Seth
LECTURES:
To gain some familiarity
with a number of different approaches to modelling and understanding adaptive processes in natural and artificial systems. In particular, to gain some understanding
of approaches to generating adaptive behaviours in autonomous robots.
PREREQUISITES
Ability to program,
familiarity with a high level procedural language. First term of EASy or
IS MSc or completion of 2nd yr. non symbolic AI course.
RATIONALE
This course will cover recent work in AI which is geared towards understanding intelligence, both in natural and artificial systems, in terms of the generation of adaptive behaviour in autonomous agents acting in dynamic uncertain environments. Adaptation will be studied at both the evolutionary and the lifetime scale.
Lectures will give
a general coverage of the area. Seminars and exercises will guide students
deeper into certain topics. Students are expected to engage in background
reading and follow-up references and techniques mentioned in the lectures.
READING LIST
There are no set texts as such. Some reading material in the form of papers will be provided. Students will find the following useful (most available in the library, bold for the best texts).
Abraham, R. and Shaw, C. 1992. Dynamics: The Geometry of Behavior. Addison-Wesley, 1992.
Arbib, M. 1989. The Metaphorical Brain 2. Wiley, 1989.
Arkin, R. Behaviour Based Robotics, MIT Press, 1998.
Ashby, W. Ross. 1960. Design for a Brain. Chapman and Hall, 1960.
Back, T. et al Handbook of Evolutionary Computing, OUP/Inst. of Physics, 1997.
Braitenberg, V. Vehicles: Experiments in Synthetic Psychology. MIT Press, 1986.
Breazeal, C. Designing Sociable Robots. MIT Press, 2004.
Brooks, R. Cambrian Intelligence: The early history of the new AI, MIT Press, 1999.
Clark, A. Being There. MIT Press, 1998.
Dayan, P. and Abbott, L. Theoretical Neuroscience. MIT Press, 2001.
Dupuy, J. The Mechanization of Mind. Princeton University Press, 2000.
Forrest, S. Emergent Computation. MIT Press, 1991.
Godfrey-Smith, P.G. Complexity and the Function of Mind in Nature. MIT Press, 1996.
Goldberg, D. Genetic Algorithms, Addison-Wesley, 1989.
Holland, J. Adaptation in natural and artificial systems .2nd ed. MIT press, 1992.
Horst-Jansen, H. Catching Ourselves in the Act. MIT Press, 1996.
Husbands, P., Holland O., and Wheeler, M. (Eds), The Mechanical Mind in History, MIT Press, 2008.
Husbands, P. and Meyer, J-A.. Evolutionary Robotics, Springer-Verlag, LNCS 1468, 1998.
Langton, C. Artificial Life: An Overview. MIT Press, 1997.
Maes, P. Designing Autonomous Agents. MIT Press, 1990.
Maynard Smith, J. Evolution and the theory of games. CUP, 1982.
Mitchell, M. An introduction to genetic algorithms MIT Press, 1996.
Nolfi, S. and Floreano, S. Evolutionary Robotics, MIT Press, 2000.
Pfeifer, R. and Bongard, J. How The Body Shapes The Way We Think. MIT Press, 2006.
Pfeifer, R. and Scheier, C. Understanding Intelligence, MIT Press, 1999.
Thrun, S., Burgard, W. and Fox, D. Probabilistic Robotics. MIT Press, 2005.
Wiener, N. Cybernetics. 2nd ed. MIT Press, 1962.
Conference Proceedings:
SABxx: From Animals to Animats: Proceedings of the International Conference on the Simulation of Adaptive Behavior (1990-)
ALIFExx: Artificial Life: Proceedings of the International Conference on Artificial Life (1987-)
ECALxx: European Conference on Artificial Life (1989-)
Journals:
Adaptive Behavior, Artificial Life, Evolutionary Computation, Cognitive Neurodynamics, Cognitive Computation, Phenomenology and the Cognitive Sciences, and increasingly many others as well.
See also the wide range of publications available from the CCNR site.
A number of
web
resources, including online reading material and details about assessment.
LECTURE SCHEDULE (TENTATIVE)
The lectures listed
below will not necessarily take exactly one fifty minute period each. Copies
of lecture slides will be made available online. This list is
tentative.
Lecture schedule still in preparation.
ASSESSMENT
N.B. Assessment
modes depend on whether you are taking this course as an undergraduate
or a postgraduate.
Undergraduates
Course assessment is split into two parts, with the weights indicated.
Exercise 1 (50%).
Programming exercise.
You will be given a choice between a GA project and a robot programming
project. A 2,000 word program report is to be handed into School office
by 4.00 pm on Thursday, week 1 of Summer Term.
Exercise 2 (50%).
3,000 word essay
on a relevant subject of your choice based on the seminars (to be agreed
with Prof. Holland by end of week 8 -- suggested titles will be made
available). To be handed in to School Office by 4.00 pm on Thursday, week 1
of Summer Term.
Consult your handbook
for penalties for late submissions. You should plan to complete your
exercises early, since last minute machine downtime or overloading will
not be taken into account. A list of suitable essay topics and programming
exercise will be made available. It is advised that you seek feedback
on your choice by week 7 by presenting a short proposal for each exercises.
Postgraduates
You will be assessed by a 5,000 word term paper (100%) based on a programming or robotic project (and containing essay elements as well) to be handed into School office by 12.00 pm on first day of Summer term. Suitable topics for programming projects will be made available, but it is expected and advised that you should be able to extend on them and come up with your own ideas. It is advised that you present a short proposal by week 7.
Refer to Handbook
for Candidates for MSc degrees for regulations on late submissions.
SEMINARS/LAB CLASS
There will be three
seminars and a lab. class during the course. The class
will be split into groups for these. Seminars will take place in weeks
3, 4, 5 and 6 only. The lab. class will take place in week 7 for PGs. You will be notified of any changes.
Reading material will be made available, watch out for future announcements.