Autonomous acquisition of the meaning of sensory states through sensory-invariance driven action
Academic Article
Overview
Research
Identity
Additional Document Info
Other
View All
Overview
abstract
How can artificial or natural agents autonomously gain understanding of its own internal (sensory) state? This is an important question not just for physically embodied agents but also for software agents in the information technology environment. In this paper, we investigate this issue in the context of a simple biologically motivated sensorimotor agent. We observe and acknowledge, as many other researchers do, that action plays a key role in providing meaning to the sensory state. However, our approach differs from the others: We propose a new learning criterion, that of on-going maintenance of sensory invariance. We show that action sequence resulting from reinforcement learning of this criterion accurately portrays the property of the input that triggered a certain sensory state. This way, the meaning of a sensory state can be firmly grounded on the choreographed action which maintains invariance in the internal state. Springer-Verlag 2004.
