Scaffolding Networks: Incremental Learning and Teaching Through Questioning
This addresses the challenge of enabling machines to perform reasoning and understanding in a scalable and efficient manner, representing a novel paradigm rather than an incremental improvement.
The paper tackles the problem of teaching machines to understand text and reason by introducing a scaffolding network that uses incremental learning through teacher-student interactions and reinforcement learning, achieving state-of-the-art performance on synthetic and real datasets with scalable reasoning even with minimal human input.
We introduce a new paradigm of learning for reasoning, understanding, and prediction, as well as the scaffolding network to implement this paradigm. The scaffolding network embodies an incremental learning approach that is formulated as a teacher-student network architecture to teach machines how to understand text and do reasoning. The key to our computational scaffolding approach is the interactions between the teacher and the student through sequential questioning. The student observes each sentence in the text incrementally, and it uses an attention-based neural net to discover and register the key information in relation to its current memory. Meanwhile, the teacher asks questions about the observed text, and the student network gets rewarded by correctly answering these questions. The entire network is updated continually using reinforcement learning. Our experimental results on synthetic and real datasets show that the scaffolding network not only outperforms state-of-the-art methods but also learns to do reasoning in a scalable way even with little human generated input.