Learning graph structure for multi-label image classification via clique generation
- 1 June 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10636919,p. 4100-4109
Abstract
Exploiting label dependency for multi-label image classification can significantly improve classification performance. Probabilistic Graphical Models are one of the primary methods for representing such dependencies. The structure of graphical models, however, is either determined heuristically or learned from very limited information. Moreover, neither of these approaches scales well to large or complex graphs. We propose a principled way to learn the structure of a graphical model by considering input features and labels, together with loss functions. We formulate this problem into a max-margin framework initially, and then transform it into a convex programming problem. Finally, we propose a highly scalable procedure that activates a set of cliques iteratively. Our approach exhibits both strong theoretical properties and a significant performance improvement over state-of-the-art methods on both synthetic and real-world data sets.Keywords
This publication has 17 references indexed in Scilit:
- Multilabel classification through random graph ensemblesMachine Learning, 2014
- New Graph Structured Sparsity Model for Multi-label Image Annotations2013
- A Scalable Stagewise Approach to Large-Margin Multiclass Loss-Based BoostingIEEE Transactions on Neural Networks and Learning Systems, 2013
- Multilabel Classification with Principal Label Space TransformationNeural Computation, 2012
- On label dependence and loss minimization in multi-label classificationMachine Learning, 2012
- Multilabel classification via calibrated label rankingMachine Learning, 2008
- Training structural SVMs when exact inference is intractable2008
- Collective multi-label classification2005
- Approximating discrete probability distributions with dependence treesIEEE Transactions on Information Theory, 1968
- On general minimax theoremsPacific Journal of Mathematics, 1958