STAND: Self-Aware Precondition Induction for Interactive Task Learning
This addresses the challenge of enabling AI agents to learn new capabilities from limited human instruction in ITL scenarios, offering incremental improvements for more consistent training experiences.
The paper tackles the problem of data-efficient rule precondition induction in interactive task learning (ITL) by introducing STAND, a method that beats popular models like XGBoost and decision trees in small-data tasks and provides accurate metrics of training progress to users.
In interactive task learning (ITL), AI agents learn new capabilities from limited human instruction provided during task execution. STAND is a new method of data-efficient rule precondition induction specifically designed for these human-in-the-loop training scenarios. A key feature of STAND is its self-awareness of its own learning -- it can provide accurate metrics of training progress back to users. STAND beats popular methods like XGBoost, decision trees, random forests, and version spaces at small-data precondition induction tasks, and is highly accurate at estimating when its performance improves on holdout examples. In our evaluations, we find that STAND shows more monotonic improvement than other models with low rates of error recurrence. These features of STAND support a more consistent training experience, enabling human instructors to estimate when they are finished training and providing active-learning support by identifying trouble spots where more training is required.