- What is development? What conditions are necessary and
sufficient in order to call a system developmental? Can we specify a
litmus test that differentiates between a developmental and a
non-developmental system? While local interaction and unfolding over time
have been traditional hallmarks of G&DS research, are they in fact
computational burdens that should be abstracted away?
- Is the purpose
of building a developmental system to model nature or simply to abstract its essential elements in
pursuit of better algorithms?
- When studying a concrete example of natural
development, how might one decide on what is relevant to the
modeling of development? And in reverse, how might one transfer mechanisms of
development from abstract systems to physical ones?
- What is the
relationship(s) between evolution and development? Is the purpose of evolution confined to
searching for acceptable configurations of a developmental programme? How can evolution computation benefit from
development?
- Why do we need automated tools that produce complex software/hardware
artifacts/designs? Is G&DS the only branch of
Evolutionary Computation that has the potential to produce artifacts that
rival the complexity of
organisms in nature? Are there non-evolutionary and
non-developmental approaches that can produce equally complex
artifacts?
- How may we incorporate into highly-adaptable physical systems, the ‘intrinsic’
qualities of development, such as self-organization, self-replication and self-repair?
Should we expect that development within physical artificial life shall be
observable only within the confines of the physical-chemical
mechanisms found in biology?
- What are the different approaches of
Computational Development? What do they have in common
and what differentiates them from each other? What are their strengths and
weaknesses, and how can they all be improved?
- Are qualitative observations about
empirical results, such as emergence of repeating motifs, more important than
quantitative performance
comparisons? What kind
of mathematical/statistical tools are needed to describe the dynamics of
G&DS?
- On which kinds of applications are developmental
approaches most suited? Developmental
systems grow and complexify over time. This
implies that they should best be applied to problems which do not have
fixed dimensionality.