017

Authors:

 Beyls, Peter.

Title:

 Evolutionary strategies for spontaneous man-machine interaction

Keywords:

 evolution, genetic algorithms, real-time interactive music systems, software agents

Abstract:

 The present paper documents a long time project aiming the computer representation of a synthetic, expressive personality endowed with expertise to successfully support rich, man-machine conversations in the musical domain. We briefly describe the developmental history of the Oscar (acronym for Oscillator Artist) application since its first incarnation in 1986. The current object-oriented implementation incorporates ideas from knowledge based systems as well as agents inspired models. The basic idea is that Oscar tries to express its personal character while also pursuing integration into a larger social context. This results in a problem of conflict resolution i.e. the continuous evaluation of expressive versus integrative forces. The program consists of some 40 modules including, a neural network for real-time tonal inference, a complex memory management scheme, extensive feature detection and feature tracking algorithms for analysing incoming MIDI streams. Oscar uses various generative methods for producing output including a physical model acting as a complex dynamical system and an ensemble of interacting musical agents (roughly based on Minsky's SOM theory) all having access to a large library of transformers. However, two unusual additional software components provide Oscar with exceptional flexibility. First, an exploration modules which automatically searches for interesting events in long term memory (inspired by the pioneering work of Doug Lenat) and second, the possibility to evolve brains -- along the lines of current work in the discipline of artificial life. Oscar's brain is a sensor-activator network with sensors, neurons and activators configured in an arbitrary network. As we perceive it to be impossible to design brains through explicit specification -- both because their complexity is beyond human understanding and because we want to favour a computational climate aiming the discovery of useful though unknown machine interpretations -- we opt for implicit evolution as an alternative to explicit design. In practice, families of brains are evolved, evaluated and attributed a certain fitness according to how well they perform. Brains (the connections and the weighting factors) are viewed as genetic material which is manipulated using cross-over and mutation functions. The objective, of course, is to evolve artificial brains which perform well facing the unpredictable input of a human interactor. The present work provides evidence that various musical imperatives like goal achievement, free improvisation, rich forms of interaction and expression can be achieved using genetic approaches. In any case, the intention is to create a computational climate where man and machine contribute to the expression of musical ideas in a common effort while sharing responsibility. A live demo of the current implementation concludes the presentation.
Beyls, Peter. peter@arti.vub.ac.be St Lukas Hogeschool Brussels Peter Beyls has been exploring computer programming for aesthetic decision making since the early Seventies both in the realm of computer graphics and interactive computer music systems. Beyls studied electronic engineering and music in Brussels and London. Has published extensively on real-time expert systems, complex dynamics, cellular automata and genetic algorithms. Lectured at various institutions in the US and Japan. Heads the Media Unit at the Artificial Intelligence Lab of VUB for five years, now teaches aesthetics of new media and sound design at St Lukas Hogeschool Brussels. Beyls has been involved with ISEA from its inception and has served on the ISEA Board of Directors until 1998.