Lifelong Wandering: A realistic few-shot online continual learning setting
This work addresses a key challenge for robotics applications where models must learn new object classes continuously without forgetting previous ones, though it is incremental as it builds on existing few-shot and continual learning paradigms.
The paper tackles the problem of catastrophic forgetting in online few-shot learning by extending it to a more realistic continual learning setting across multiple indoor environments, showing that existing methods face a trade-off between online performance and forgetting.
Online few-shot learning describes a setting where models are trained and evaluated on a stream of data while learning emerging classes. While prior work in this setting has achieved very promising performance on instance classification when learning from data-streams composed of a single indoor environment, we propose to extend this setting to consider object classification on a series of several indoor environments, which is likely to occur in applications such as robotics. Importantly, our setting, which we refer to as online few-shot continual learning, injects the well-studied issue of catastrophic forgetting into the few-shot online learning paradigm. In this work, we benchmark several existing methods and adapted baselines within our setting, and show there exists a trade-off between catastrophic forgetting and online performance. Our findings motivate the need for future work in this setting, which can achieve better online performance without catastrophic forgetting.