Unsupervised Behavior Discovery With Quality-Diversity Optimization

Luca Grillotti

doi:10.1109/TEVC.2022.3159855

2022-12-01

PRE

Luca Grillotti *

Antoine Cully

DOI：10.1109/TEVC.2022.3159855

原文链接

原文求助

摘要

En 中文

Quality-diversity (QD) algorithms refer to a class of evolutionary algorithms designed to find a collection of diverse and high-performing solutions to a given problem. In robotics, such algorithms can be used for generating a collection of controllers covering most of the possible behaviors of a robot. To do so, these algorithms associate a behavioral descriptor (BD) to each of these behaviors. Each BD is used for estimating the novelty of one behavior compared to the others. In most existing algorithms, the BD needs to be hand-coded, thus requiring prior knowledge about the task to solve. In this article, we introduce: autonomous robots realizing their abilities, an algorithm that uses a dimensionality reduction technique to automatically learn BDs based on raw sensory data. The performance of this algorithm is assessed on three robotic tasks in simulation. The experimental results show that it performs similar to traditional hand-coded approaches without the requirement to provide any hand-coded BD. In the collection of diverse and high-performing solutions, it also manages to find behaviors that are novel with respect to more features than its hand-coded baselines. Finally, we introduce a variant of the algorithm which is robust to the dimensionality of the BD space.

Keyword:

Behavioral diversity

optimization methods

quality-diversity (QD) optimization

robotics

unsupervised learning