Tipping the scales: Guidance and Intrinsically Motivated Behavior

Abstract: We propose an approach to learning in autonomous agents that is complementary to active learning and intrinsically motivated learning as it relies on a dynamical maintenance of a controllable sensorimotor loop which is continuously modulated by weak teaching signals. In this way a symbolic representation of the task can be realized in the behavior of the robot across a variety of robot hardware parameters, environments and different levels of experience. We demonstrate the control algorithm to have good scalability with respect to the number of degrees of freedom. The control signals may be externally given but can often also be acquired by extracting regularities within the learning process.

Video 1: Armband robot learns to locomote – weakly guided. Behavior of the robot with cross-motor teaching and weak guidance (γ=0.001). A slow locomotive behavior with different velocities is exhibited. Explorative actions cause the posture of the robot to vary in the course of time.

Video 2: Armband robot quickly learns to locomote. Behavior of the robot with cross-motor teaching and medium guidance (γ=0.003). Comparable fast locomotive behavior emerges quickly and is persistent. Nevertheless the velocity varies. Only small exploratory actions are takes, such that the posture is mainly constant.

Video 3: Armband changing the direction of motion. The behavior of the robot with cross-motor teaching when the connections are changed. The video start with a fast locomotive behavior to the left (k=1). At time 5:00 the couplings are changed (k=0) and the robot slowly stops. A period of probing actions follows until a reversed locomotion starts to show up.

This document was translated from L^AT_EX by H^EV^EA.