Learning to Learn and Predict: A Meta-Learning Approach for Multi-Label Classification
This work addresses multi-label classification tasks in natural language processing, offering an incremental improvement by incorporating label dependencies through meta-learning.
The paper tackles the problem of multi-label classification by addressing the complexity and dependencies among labels, proposing a meta-learning method that learns training and prediction policies for each label, resulting in more accurate classification results on fine-grained entity typing and text classification tasks.
Many tasks in natural language processing can be viewed as multi-label classification problems. However, most of the existing models are trained with the standard cross-entropy loss function and use a fixed prediction policy (e.g., a threshold of 0.5) for all the labels, which completely ignores the complexity and dependencies among different labels. In this paper, we propose a meta-learning method to capture these complex label dependencies. More specifically, our method utilizes a meta-learner to jointly learn the training policies and prediction policies for different labels. The training policies are then used to train the classifier with the cross-entropy loss function, and the prediction policies are further implemented for prediction. Experimental results on fine-grained entity typing and text classification demonstrate that our proposed method can obtain more accurate multi-label classification results.