Information Technology Reference
In-Depth Information
Fig. 12.1
Artists (and observers of art) get intrinsically rewarded for making (and observing)
novel patterns: data that is neither arbitrary (like incompressible random white noise) nor regular
in an already known way, but regular in a way that is new with respect to the observer's current
knowledge, yet learnable. While the Formal Theory of Creativity explains the desire to create or
observe all kinds of art, low-complexity art (Schmidhuber
1997c
) illustrates it in a particularly
clear way. Many observers report they derive pleasure or aesthetic reward from discovering simple
but novel patterns while actively scanning the self-similar
Femme Fractale
above (Schmidhuber
1997b
). The observer's learning process causes a reduction of the subjective compressibility of
the data, yielding a temporarily high derivative of subjectively perceived simplicity or elegance or
beauty: a temporarily steep learning curve. The corresponding intrinsic reward motivates him to
keep looking at the image for a while. Similarly, the computer-aided artist got reward for discov-
ering a satisfactory way of using fractal circles to create this low-complexity artwork, although it
took him a long time and thousands of frustrating trials. Here is the explanation of the artwork's
low algorithmic complexity: The frame is a circle; its leftmost point is the centre of another circle
of the same size. Wherever two circles of equal size touch or intersect are centres of two more
circles with equal and half size, respectively. Each line of the drawing is a segment of some cir-
cle, its endpoints are where circles touch or intersect. There are few big circles and many small
ones. This can be used to encode the image very efficiently through a very short program. ©Jürgen
Schmidhuber, 1997-2010
12.6 Low-Complexity Art as End Product of a Search Process
Modelled by the Formal Theory of Creativity
Low-complexity art (Schmidhuber
1997c
) may be viewed as the computer-age
equivalent of minimal art. To depict the essence of objects, it builds on concepts
from algorithmic information theory (Solomonoff
1978
, Kolmogorov
1965
, Li and
