Collaboration based Multi-Label Learning
This work addresses the challenge of accurately modeling label correlations for multi-label learning tasks, which is incremental as it builds on existing methods by explicitly incorporating correlations into predictions.
The paper tackles the problem of exploiting label correlations in multi-label learning by proposing a method that learns correlations via sparse reconstruction and integrates them into model training to achieve correlated predictions. Experimental results show the approach outperforms state-of-the-art methods.
It is well-known that exploiting label correlations is crucially important to multi-label learning. Most of the existing approaches take label correlations as prior knowledge, which may not correctly characterize the real relationships among labels. Besides, label correlations are normally used to regularize the hypothesis space, while the final predictions are not explicitly correlated. In this paper, we suggest that for each individual label, the final prediction involves the collaboration between its own prediction and the predictions of other labels. Based on this assumption, we first propose a novel method to learn the label correlations via sparse reconstruction in the label space. Then, by seamlessly integrating the learned label correlations into model training, we propose a novel multi-label learning approach that aims to explicitly account for the correlated predictions of labels while training the desired model simultaneously. Extensive experimental results show that our approach outperforms the state-of-the-art counterparts.