Eytan Ruppin ( ruppin@math.tau.ac.il ) 
2st Semester, 1998/99 - Tuesday 14-16  
School of Mathematical Sciences, 
Tel-Aviv University

Seminar on Evolutionary Autonomous Agents (Alife) - 0368.4050.01

Text book:

A few textbooks may serve as general background for the seminar:
Introduction to Genetic Algorithms, by Melanie Mitchell, Santa Fe Inst., 1995.
Artificial Life, by Christopher Langton, 1995.
Artificial Life, by David Fogel, 1995.
(Copies of all these books are held by the library of exact sciences).

Seminar syllabus:

After a brief introduction to alife and genetic algorithms, we shall focus on the computational evolution of autonomous agents. More specifically, we shall concentrate on neural-network guided agents, and mainly on such software agents (``softbots''). Our goal is to get a feeling of what can be currently achieved with such agents, what tasks can they perform and what it takes to successfully evolve them. Issues addressed will include encoding schemes, the interaction of learning and evolution, and `higher-level' functions. Finally, several other `applications' of alife to computer science (other than autonomous agents) will be discussed.

List of tentative papers to be covered:

Enclosed below is the list of tentative papers for the seminar. The abstracts of these papers may be seen by locating the pertaining journal vi the link to Complex Adaptive Systems and Artificial Life in my homepage.
  • 1. The perfect C. Elegans project, Artificial Life, 4(2), 141-156, 1998.
  • 2. Self adaptation in evolving systems, Artificial Life, 4(2), 183-201, 1998.
  • 3. PtGAs -- Genetic algorithms evolving non-coding segments by means of promoter/terminator sequences, Evolutionary Computation, 6(4), 361-386, 1998.
  • 4. Conjugate schema and basis representation of crossover and mutation operators, Evolutionary Computation, 6(2), 129-160, 1998.
  • 5. Adapting operator settings in GAs, Evolutionary Computation, 6(2), 161-184, 1998.
  • 6. Evolutionary induction of sparse neural trees, Evolutionary Computation, 5(2), 213-236, 1997.
  • 7. Changing representation during search: a comparative study of delta coding, Evolutionary Computation, 2(3), 249-278, 1994.
  • 8. Evolutionary robotics and the radical envelope-of-noise hypothesis, Adaptive Behavior, 6(2), 325-368, 1997.
  • 9. Automatic definition of modular NNs, Adaptive Behavior, 3(2), 151-183, 1994.
  • 10. Sequential behavior and learning in evolved dynamical neural networks, Adaptive Behavior, 2(3), 219-246, 1993.
  • 11. On the dynamics of small continuous-time recurrent neural networks, Adaptive Behavior, 3(4), 469-509, 1994.
  • 12. Dynamical modules in an evolved model CPG for walking, Journal of Neurophysiology ? [Probably may be down-loaded from Randolph Beer's homepage at ???].
  • 13. Neural Network representation using Kauffman Networks, Artificial Life, 3(2), 67-80, 1997.
  • 14. Using emergent modularity to develop control systems for mobile robots, Adaptive Behavior, 5(3,4), 343-364, 1996.
  • 15. Incremental evolution of complex general behavior, Adaptive Behavior, 5(3,4), 317-342, 1996.

    • Note: Participation in the seminar this year will be strictly limited to twnety people. There will be vanishing overlap with the papers covered in the seminar last year.
    Last updated April, 1999