The traditional tness function based methodology of artiicial evolution is argued to be inadequate for the construction of entities with behaviors novel to their designers. Evolutionary emergence via natural selection (without an explicit tness function) is the way forward. This paper further considers the question of what to evolve, the focus being on principles of developmental modularity in neural networks. To develop and test the ideas, an artiicial world containing autonomous organisms has been created and is described. Results show the developmental system to be well suited to long-term in-cremental evolution. Novel emergent strategies are iden-tiied both from an observer's perspective and in terms of their neural mechanisms. The Artiicial Life goal presents us with the problem that we do not understand (natural) life well enough to specify it to a machine. Therefore we must either increase our understanding of it until we can, or create a system which outperforms the speciications we can give it. The rst possibility includes the traditional top-down methodology, which is clearly as inappropriate for ALife as it has proved to be for AI. It also includes manual in-cremental (bottom-up) construction of autonomous systems with the aim of increasing our understanding and ability to model life by building increasingly impressive systems, retaining functional validity by testing them within their destination environments. The second option is to create systems which out-perform the speciications given them and which are open to producing behaviors comparable with those of (albeit simple) natural life. Evolution in nature has no (explicit) evaluation function. Through organism-environment interactions, including interactions between similarly-capable organisms, certain behaviors fare better than others. This is how the non-random cumulative selection works without any long-term goal. It is why novel structures and behaviors emerge. As artiicial evolution is applied to increasingly complex problems, the diiculty in specifying satisfactory evaluation functions is becoming apparent { see (Zaera, Clii & Bruten 1996), for example. At the same time, the power of natural selection is being demonstrated in prototypal systems such as Tierra (Ray 1991) and Poly-World (Yaeger 1993). Artiicial selection involves the imposition of an artiice crafted for some cause external to a system beneath it while natural selection does not. Natural selection is necessary for evolutionary emergence but does not imply sustained emergence (evermore new emergent phenomena) and the question \what should we evolve?" needs to be answered with that in mind (Chan-non & Damper 1998). This paper sets …
[1]
Inman Harvey,et al.
Incremental evolution of neural network architectures for adaptive behavior
,
1993,
ESANN.
[2]
L. Yaeger.
Computational Genetics, Physiology, Metabolism, Neural Systems, Learning, Vision, and Behavior or PolyWorld: Life in a New Context
,
1997
.
[3]
Thomas S. Ray,et al.
An Approach to the Synthesis of Life
,
1991
.
[4]
Robert I. Damper,et al.
Perpetuating evolutionary emergence
,
1998
.
[5]
Stewart W. Wilson,et al.
Not) Evolving Collective Behaviours in Synthetic Fish
,
1996
.
[6]
Inman Harvey,et al.
Evolutionary robotics and SAGA: The case for hill crawling and tournament selection
,
1994
.
[7]
Hiroaki Kitano,et al.
Designing Neural Networks Using Genetic Algorithms with Graph Generation System
,
1990,
Complex Syst..
[8]
Aristid Lindenmayer,et al.
Mathematical Models for Cellular Interactions in Development
,
1968
.
[9]
Frédéric Gruau.
Artificial Cellular Development in Optimization and Compilation
,
1995,
Towards Evolvable Hardware.
[10]
Vidroha Debroy,et al.
Genetic Programming
,
1998,
Lecture Notes in Computer Science.