A neural architecture for autonomous learning

Abstract

Psychology and neurobiology nowadays provide a large amount of precise information on visual system function. This information can be used in the design of autonomous systems capable of learning and recognising objects and places important for survival in complex unknown (real or virtual) environments. Our work is based on the principles that perception is fundamentally a dynamic process in constant interaction with movement; and that learning can be made simpler if the systems are not required to learn the invariants of their environment (e.g. preservation of neighbour topological relations, or connectivity of the space). The techniques that contribute to devising these adaptive systems in continuous interaction with their environment could significantly influence our approach to programming and the man‐machine interface.

Citation

Gaussier, P., Joulain, C., Banquet, J.P., Revel, A., Lepretre, S. and Moga, S. (1999), "A neural architecture for autonomous learning", Industrial Robot, Vol. 26 No. 1, pp. 33-38. https://doi.org/10.1108/01439919910250205