Results 1 -
2 of
2
Integrating Reinforcement-Learning and Accumulator Models to Study Decision Making and Reaching Behaviours
"... This paper presents a model of organisms ’ sensorimotor system based on the idea that complex behaviors are build on the basis of repertoires of sensorimotor primitives organised around specific goals (in this case, arm postures). The architecture of the model incorporates an actor-critic reinforcem ..."
Abstract
-
Cited by 1 (1 self)
- Add to MetaCart
This paper presents a model of organisms ’ sensorimotor system based on the idea that complex behaviors are build on the basis of repertoires of sensorimotor primitives organised around specific goals (in this case, arm postures). The architecture of the model incorporates an actor-critic reinforcement learning system, enhanced with an “accumulator model ” for decision making, capable of selecting sensorimotor primitives to accomplish a discrimination reaching task that has been used in physiological studies of monkeys ’ premotor cortex. The results show that the proposed architecture is a first important step towards the construction of a biologically-sound integrated model of the primitive-based hierarchical organization of organisms ’ sensorimotor systems.
