papers AI Learner
The Github is limit! Click to go to the new site.

Skill Acquisition via Automated Multi-Coordinate Cost Balancing

2019-03-27
Harish Ravichandar, S. Reza Ahmadzadeh, M. Asif Rana, Sonia Chernova

Abstract

We propose a learning framework, named Multi-Coordinate Cost Balancing (MCCB), to address the problem of acquiring point-to-point movement skills from demonstrations. MCCB encodes demonstrations simultaneously in multiple differential coordinates that specify local geometric properties. MCCB generates reproductions by solving a convex optimization problem with a multi-coordinate cost function and linear constraints on the reproductions, such as initial, target, and via points. Further, since the relative importance of each coordinate system in the cost function might be unknown for a given skill, MCCB learns optimal weighting factors that balance the cost function. We demonstrate the effectiveness of MCCB via detailed experiments conducted on one handwriting dataset and three complex skill datasets.

Abstract (translated by Google)
URL

http://arxiv.org/abs/1903.11725

PDF

http://arxiv.org/pdf/1903.11725


Comments

Content