Two-view feature generation model for semi-supervised learning
Rie Kubota Ando, Tong Zhang · International Conference on Machine Learning · 2007
We consider a setting for discriminative semi-supervised learning where unlabeled data are used with a generative model to learn effective feature representations for discriminative training. Within this framework, we revisit the two-view feature generation model of co-training and prove that the optimum predictor can be expressed as a linear combination of a few features constructed from unlabeled data. From this analysis, we derive methods that employ two views but are very different from co-training. Experiments show that our approach is more robust than co-training and EM, under various data generation conditions.
Cited by 90 works. doi:10.1145/1273496.1273500
Authors
References
- Statistical Learning Theory
- Combining labeled and unlabeled data with co-training
- Text Classification from Labeled and Unlabeled Documents using EM
- Semi-supervised learning using Gaussian fields and harmonic functions
- A Framework for Learning Predictive Structures from Multiple Tasks and Unlabeled Data
- A probability analysis on the value of unlabeled data for classification problems
- PAC Generalization Bounds for Co-training