At the moment, deep learning techniques have helped systems learn to solve complex tasks in a way that more closely matches what humans can do—but those tasks are still specific, such as beating a human at a game. And they require a rigid sequence: gather data, determine the goal, deploy an algorithm.

This process requires humans and can be time-consuming, especially during early phases when supervised training is required. Continuous learning (CL) is more about autonomous and incremental skill building and development, and researchers will continue to push the limits of what’s possible in this field.