Physical Models that Learn

Nicolas Szilas (1, 2) , Claude Cadoz (1, 2)

1 ACROE - ACROE
2 LIFIA - Laboratoire d'Informatique Fondamentale et d'Intelligence Artificielle

In this paper is proposed an analysis method for physical models networks, based on connectionist learning algorithms. The first experimental results, obtained on limited cases, are quite encouraging and suggest dome improvements, using the interactivity of Physical Models.

Domaines: Modélisation et simulation; Informatique

Nicolas Szilas, Claude Cadoz. Physical Models that Learn. International Computer Music Conference 1993, 1993, Waseda, Japan. pp.4. ⟨hal-00910607⟩