Full professor of Artificial Intelligence
School of Computer Science - University of Malaga
Technology Park of Andalusia
Severo Ochoa, 4 - 107 29590 Malaga (Spain)
After dedicating the first 15 years of my research activity to brain function modeling, my interests steered towards the understanding of life's complexity. Topics like the emergence of living matter, and the evolution of complex forms and behaviors have always been in my mind. Now we approach them with computer modeling and simulation.
Melomics (the genomics of melodies) is among our most ambitious projects. Melomics is now disclosing the biggest repository of music in the world, and applications around it that will change the way music is composed and consumed.
You may wonder how computer scientists ended up creating computers that compose professional music with no human intervention. This is how the story goes: We design Artificial Life, i.e. creatures that resemble features of leaving beings, which emerge from complex genomes, interact with artificial worlds, and might compete, cooperate or recombine with others to better adapt. By encoding complex worlds, populations of creatures evolve, eventually solving concrete problems. See below a lander which, after an evolution process, deploys its structure from an initial cell, and lands in a way which has been "sculpted" by the selection pressure.
Creatures can also adapt themselves to follow a path, finding very diverse and original behaviors, which can then be applied to automate processes in the animation industry (see at the end of this video how simple maneuvering strategies that were obtained can be combined to mimic an intelligent behavior).
Molecular motors are another example of complex structures that evolve in our computers, emerging in a world where skilled crawling along a segment is favored.
Having seen this, it's no surprise that they can also grasp the beauty, coherence and complexity that first-shelf artists, like the London Symphony Orchestra, would demand in their repertoire.
And it's not only that complex objects can emerge from new evolving realms, the point is that the genetic process is under control, giving the opportunity of mastering new forms of genetic engineering for designing physical or abstract objects. Imagine a plausible future where musicians compose by altering genetic information, designers interacte with new forms mutating actual mock-ups, animators browse completely new previews of a scene... All in all, computers truly assuming creative tasks, and augmenting human ingenuity.