Global Safe Sequential Learning via Efficient Knowledge Transfer
This work addresses the challenge of global safe exploration in sequential learning for applications with unknown safety constraints, offering a novel transfer-based approach that improves efficiency and scope over existing methods.
The paper tackles the problem of safe sequential learning being limited to local exploration around initial data, which slows task learning and leaves disconnected safe regions unexplored. The proposed safe transfer sequential learning method accelerates task learning and expands explorable safe regions by leveraging offline data from a related source task, achieving lower data consumption and enhanced global exploration across disjoint safe regions compared to state-of-the-art methods.
Sequential learning methods, such as active learning and Bayesian optimization, aim to select the most informative data for task learning. In many applications, however, data selection is constrained by unknown safety conditions, motivating the development of safe learning approaches. A promising line of safe learning methods uses Gaussian processes to model safety conditions, restricting data selection to areas with high safety confidence. However, these methods are limited to local exploration around an initial seed dataset, as safety confidence centers around observed data points. As a consequence, task exploration is slowed down and safe regions disconnected from the initial seed dataset remain unexplored. In this paper, we propose safe transfer sequential learning to accelerate task learning and to expand the explorable safe region. By leveraging abundant offline data from a related source task, our approach guides exploration in the target task more effectively. We also provide a theoretical analysis to explain why single-task method cannot cope with disconnected regions. Finally, we introduce a computationally efficient approximation of our method that reduces runtime through pre-computations. Our experiments demonstrate that this approach, compared to state-of-the-art methods, learns tasks with lower data consumption and enhances global exploration across multiple disjoint safe regions, while maintaining comparable computational efficiency.