An Exhibition about the Impact of Self-Organisation and Chaos Theory on Artificial Intelligence

On this page we will allow you to digitally explore the exhibition that was held from the 18th to the 23rd of December 2023 at the Koninklijke Bibliotheek, the Bibliothèque Royal of Brussels. Please travel across the different rooms, a plan show the room's organization and a slideshow can be played with to see the different artifacts that populated the exhibition.

Curator: Luc Steels

Technical director: Guido Lucassen
Production: Arielle Sleutel
Artworks: Anne Marie Maes
Video sources: Europace 2000

Picture: The Royal Libray of Brussels

This pop up exhibition shows videos of interviews, talks, and panels with the scientists who developed these original ideas. The exhibition also documents the impact on Artificial Intelligence in the 90s, particularly through the revivial of neural networks and efforts to build animats - autonomous robots inspired by animals. 

The exhibition is partly based on video materials coming from the television series ‘Science on the edge of Chaos’ created by Luc Steels, recorded and edited at the AV Services of the KU Leuven and broadcast by the Europace educational network operating from 1988 until 1996.

The research work is complemented with scientific classics from the KBR archives and artworks.

Room 1: PHYSICS & CHEMISTRY

In the 1950s, chemists discovered that matter may become active in far-from-equilibrium conditions. Various kinds of structures can emerge, such as oscillations or geometrical dynamical patterns. They are the outcome of local interactions between elements but nevertheless give rise to remarkable global effects. The Brussels chemist Ilya Prigogine (Nobel prize 1977) and his formidable group at the ULB in Brussels, investigated these phenomena and developed new mathematical tools that lead to a new understanding of Nature.

This section shows both natural chemical experiments and computer models of self-organizing systems as well as dialogs with Ilya Prigogine, Hermann Haken, Heinz-Otto Peitgen and Ian Stewart, and a visualization of the fractal formed with the Mandelbrot set. 

Room 2: MATHEMATICS & COMPUTING

The study of self-organizing processes requires novel mathematical tools as extensions of dynamical systems theory and geometry. The tools took the form of chaos theory and fractal geometry introducing the concepts of sensitivity to initial conditions, strange attractors, and self-similar geometric objects known as fractals.

This section shows first the work of Jacques Laskar on exploring whether the solar system exhibits chaos and therefore cannot be predicted the way that Newton or Laplace thought. The section then shows explanations of some of the most fundamental notions of chaos theory. These explanations are given by originators of these ideas: David Ruelle, Floris Takens, Yves Pomeau, Michel Hénon and Benoit Mandelbrot. Finally, this section also shows the Lorenz attractor and how two very close trajectories deviate exponentially in the long run.

Room 3: BIOLOGY

Insights into self-organisation and non-equilibrium complex dynamical systems started to make major impacts on various areas biology in the 1980s and 1990s. First, for understanding various organisms in the body that rely on dynamical systems, such as the beating of the heart. Second, for understanding the origins of life, in particular, for understanding the conditions for the arrival of the fittest, and these include self-organisation and emergent behaviours.

Third, for understanding pattern formation during development, such as the formation of the stripes on a zebra, or the colour patches on a butterfly’s wings. They are only very partially determined by genetic processes. And finally for understanding the collective behaviour of organisms, such as the aggregation of unicellular organisms into multi-cellular organisms, or the formation of intricate emergent structures such as honeycombs by bees, dams by beavers, paths by ants, or  swarms by birds.

All these examples are illustrated here with images and videos of scientists studying living systems from the angle of self-organisation against the background of classical texts in biology on the study of animals and plants.

Room 4: ART

The beauty of fractal structures and objects created by living systems through self-organization has for centuries been an important source of inspiration for artists. The illustrations we find in the earliest books on plants and animals are among the most remarkable and detailed drawings of such structures.

Also more recently, artists working on the borderlines between science and art have been making art works that mimic self-organizing processes in order to generate novel aesthetic forms or to make hidden natural behaviours and structures visible.

This exhibition features examples from the ‘Wunderkammer’ of artist Anne Marie Maes. The artworks are scattered throughout the exhibition and some more examples are shown in this section, focusing in particular on the collective behaviour of bees and the activities of bacteria while growing skins.

Room 5: NEUROSCIENCE

Since the 1980s, progress in complex systems science and chaos theory also lead to new ways of thinking about how the brain works. Although neural networks were already devised in the 1950s and studied in the 1960s and 1970s, it is only in the 1990s that they came back in the spotlight thanks to increased computer power and renewed attention to multi-layered networks and reinforcement learning.

This section starts with the original anatomical works by  Vesalius and Palfin from the 16th century that caused a revolution in the study of the brain, even though nobody could even imagine what that organ was doing. It then presents a trends in neuroscience pioneered by Francisco Varela, known as enactivism. The final section explains the foundations of neural networks as it was understood in the 1990s.

Room 6: ARTIFICIAL LIFE

In 1991 Paul Bourgine and Francisco Varela defined Artificial Life thus:

“Artificial Life embodies a recent and important conceptual step in modern science: asserting that the core of intelligence and cognitive abilities is the same as the capacity of living. … What needs to be understood and characterized is the class of processes that endow living creatures with their characteristic autonomy, key properties such as viability, abduction and adaptibility. The autonomy of living beings is understood here both with regards to their actions and to the way in which they shape their world into significance. This exploration goes hand in hand with the theory, design, and construction of simple autonomous agents. [Proceedings of ECAL, Paris, 1991]

Room 7: ANIMATS

In the early 1990s there was a wave of interest and enthusiasm for building small but agile autonomous robots packed with batteries, sensors, actuators, DIY electronics, and a processing board powerful enough to  sustain complex computations and reactive behavior in real-time. These robots were called ‘animats’ or ‘artificial creatures’. This section gives an unusual peek into a springschool where the community building these animats congregated. The school was entitled ‘The Biology and Technology of Autonomous Agents’, organised in  March 1993 by Luc Steels (VUB AI Lab) and Rodney Brooks (MIT AI Lab) in  Castle Ivano located in the Dolomites near Trento. The videos show the different ateliers, experiments, lectures, and gatherings. This section also shows how this research line developed further into the late 2000s towards humanoid robots that autonomously build shared categories and symbolic communication systems.

Lectures and social event