Adaptive tutoring improves learning by choosing best teaching strategy
AdaTutor: Adaptive Tutoring through Pedagogical Strategy Selection
Artificial IntelligenceComputation and LanguageMachine Learning
Summary
Helping students learn is not just about what to teach but also how to teach it. The authors developed a method that learns the best way to respond when a student makes mistakes, picking different teaching styles as it goes. They tested this approach on math and science problems and found it works better than always using the same teaching style. This smart tutor can also be used with other tutoring tools to improve their advice.
What this means in practice
- •For educational software developers: Create tutoring systems that dynamically pick the best response style for each student mistake, improving learning outcomes.$Commercial implications: Enables development of smarter tutoring apps that adjust feedback methods, offering personalized learning experiences for commercial edtech products.
- •For customer support platform builders: Incorporate adaptive response strategies that choose the best way to assist users based on problem type and progression.
Authors
Rajeev Bhatt Ambati, Shashank Srivastava, Snigdha Chaturvedi
Abstract
Effective tutoring requires determining not only what information to provide, but also which form of assistance to deliver. A tutor may address the same student error using various pedagogical strategies, adapting its approach as the interaction progresses. We introduce Comparative Pedagogical Supervision (CPS), a training framework to learn to select the optimal strategy and feedback for student learning. We evaluate CPS on math and science reasoning datasets. After five tutoring turns, CPS outperforms the strongest fixed-strategy baselines. The learned tutor transfers across datasets, and its strategy selection can also inform a separate API-based tutor.