发布时间：

2026-06-30

DOI码：

10.1007/s10489-023-05044-y

所属单位：

人工智能教育学部

论文名称：

Actively learning dynamical systems using Bayesian neural networks

发表刊物：

Applied Intelligence

刊物所在地：

国外学术期刊

关键字：

Active learning; Dynamical system; Bayesian neural network; Model predictive control

摘要：

Learning dynamical systems in a sample-efficient way is important for model-based control. Active learning which sequentially selects the most informative data to sample is capable of greatly reducing sample complexity. The active learning problem for dynamical systems is hard as we can not arbitrarily draw samples from the system's state space under constraints of system dynamics. The existing approaches model the dynamical systems using Bayesian linear regression or Gaussian processes which can not be applied to complex dynamical systems with high-dimensional state spaces. In this article, we propose a new method to actively learn dynamical systems using Bayesian neural networks which allow for modeling high-dimensional systems with complex dynamics. By maximizing the accumulated differential entropies along the trajectory, the proposed method iteratively searches for the most informative action sequence which will yield informative samples when applied to the real system. With random exploration and model-based reinforcement learning as baselines, we verify the superiority of the proposed method via accuracy of one-step and multi-step predictions, the control performance, the exploration efficiency of the state space on numerical benchmarks.

合写作者：

Fujimoto, Kenji,Maruta, Ichiro

第一作者：

汤胜兵

论文类型：

期刊论文

论文编号：

1732361377114361856

文献类型：

00001

ISSN号：

0924-669X

是否译文：

否

发表时间：

2023-10-27

收录刊物：

其他（ESI、Scopus、OA）、EI、SCI