Knowing when we do not know: Bayesian continual learning for sensing-based analysis tasks
This work addresses a more realistic aspect of continual learning for sensing-based analysis tasks, though it appears incremental as it builds on existing Bayesian methods without introducing a new paradigm.
The paper tackles the problem of continual learning in realistic scenarios where some tasks are more critical to remember than others, proposing a Bayesian inference framework that can prioritize remembering old tasks or learning new ones, and demonstrates robustness in adapting to changing sensing environments while using prediction uncertainty to assess reliability.
Despite much research targeted at enabling conventional machine learning models to continually learn tasks and data distributions sequentially without forgetting the knowledge acquired, little effort has been devoted to account for more realistic situations where learning some tasks accurately might be more critical than forgetting previous ones. In this paper we propose a Bayesian inference based framework to continually learn a set of real-world, sensing-based analysis tasks that can be tuned to prioritize the remembering of previously learned tasks or the learning of new ones. Our experiments prove the robustness and reliability of the learned models to adapt to the changing sensing environment, and show the suitability of using uncertainty of the predictions to assess their reliability.