Curriculum Abductive Learning
This addresses training challenges in integrating machine learning with logical reasoning, particularly for complex knowledge bases, but is incremental as it builds on prior ABL methods.
The paper tackles the instability and large abduction space in Abductive Learning (ABL) by proposing Curriculum Abductive Learning (C-ABL), which partitions the knowledge base into sub-bases introduced progressively during training, resulting in improved training stability, convergence speed, and final accuracy across multiple tasks.
Abductive Learning (ABL) integrates machine learning with logical reasoning in a loop: a learning model predicts symbolic concept labels from raw inputs, which are revised through abduction using domain knowledge and then fed back for retraining. However, due to the nondeterminism of abduction, the training process often suffers from instability, especially when the knowledge base is large and complex, resulting in a prohibitively large abduction space. While prior works focus on improving candidate selection within this space, they typically treat the knowledge base as a static black box. In this work, we propose Curriculum Abductive Learning (C-ABL), a method that explicitly leverages the internal structure of the knowledge base to address the ABL training challenges. C-ABL partitions the knowledge base into a sequence of sub-bases, progressively introduced during training. This reduces the abduction space throughout training and enables the model to incorporate logic in a stepwise, smooth way. Experiments across multiple tasks show that C-ABL outperforms previous ABL implementations, significantly improves training stability, convergence speed, and final accuracy, especially under complex knowledge setting.