Open-World Continual Learning: A Framework

As more and more AI agents are used in practice, we need to think about how to make these agents fully autonomous so that they can (1) learn by themselves continually in a self-motivated and self-initiated manner rather than being retrained offline periodically on the initiation of human engineers and (2) accommodate or adapt to unexpected or novel circumstances. As the real-world is an open environment that is full of unknowns or novelties, detecting novelties, characterizing them, accommodating or adapting to them, and gathering ground-truth training data and incrementally learning the unknowns/novelties are critical to making AI agents more and more knowledgeable and powerful over time. This chapter presents a theoretical framework for this new paradigm. The framework is called Self-initiated Open-world continual Learning and Adaptation (SOLA).