|
|
Natural systems process information in
ways that are unrivaled by current computers. They learn
to recognize relevant patterns, they remember patterns
that have been seen previously, construct internal models
of their environments, use combinatorics effectively, and
their data and control structures are massively parallel
and highly distributed. Examples of such systems are widespread
in biology (including neuroscience, vision, immunology and
ecology), physics, chemistry, and social systems.
We seek both to
understand these natural processes and to apply their
organizing principles in computation. The adaptive
computation group at UNM has a
special interest in
interactions between biology and computation. We believe that modern
computer systems have many properties in common with living systems
and should be engineered accordingly. Thus, we emphasize concrete
implementations of our ideas, building real
systems, and solving real computer systems problems. Modeling
projects in the group range from abstract "artificial-life" style
models that capture qualitative properties to much more detailed
models based on real data that make quantitative predictions.
The group has close ties with UNM's Biology Department
and the Santa Fe
Institute.
|