Semi-supervised classification with hybrid generative/discriminative methods
Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining
Classification using discriminative restricted Boltzmann machines
Proceedings of the 25th international conference on Machine learning
An asymptotic analysis of generative, discriminative, and pseudolikelihood estimators
Proceedings of the 25th international conference on Machine learning
The asymptotics of semi-supervised learning in discriminative probabilistic models
Proceedings of the 25th international conference on Machine learning
Mining the web for visual concepts
Proceedings of the 9th International Workshop on Multimedia Data Mining: held in conjunction with the ACM SIGKDD 2008
An improved hierarchical Bayesian model of language for document classification
COLING '08 Proceedings of the 22nd International Conference on Computational Linguistics - Volume 1
Learning Deep Architectures for AI
Foundations and Trends® in Machine Learning
Exponential family hybrid semi-supervised learning
IJCAI'09 Proceedings of the 21st international jont conference on Artifical intelligence
Semi-supervised learning of visual classifiers from web images and text
IJCAI'09 Proceedings of the 21st international jont conference on Artifical intelligence
Help-training semi-supervised LS-SVM
IJCNN'09 Proceedings of the 2009 international joint conference on Neural Networks
Learning Visual Object Categories for Robot Affordance Prediction
International Journal of Robotics Research
Object classification by fusing SVMs and Gaussian mixtures
Pattern Recognition
Why Does Unsupervised Pre-training Help Deep Learning?
The Journal of Machine Learning Research
A robust semi-supervised classification method for transfer learning
CIKM '10 Proceedings of the 19th ACM international conference on Information and knowledge management
Help-Training for semi-supervised support vector machines
Pattern Recognition
Semi-supervised latent variable models for sentence-level sentiment analysis
HLT '11 Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies: short papers - Volume 2
Information, Divergence and Risk for Binary Experiments
The Journal of Machine Learning Research
Locally discriminative topic modeling
Pattern Recognition
Aspects of semi-supervised and active learning in conditional random fields
ECML PKDD'11 Proceedings of the 2011 European conference on Machine learning and knowledge discovery in databases - Volume Part III
Hybrid generative-discriminative nucleus classification of renal cell carcinoma
SIMBAD'11 Proceedings of the First international conference on Similarity-based pattern recognition
Renal cancer cell classification using generative embeddings and information theoretic kernels
PRIB'11 Proceedings of the 6th IAPR international conference on Pattern recognition in bioinformatics
Pattern Recognition Letters
Multi-class particle swarm model selection for automatic image annotation
Expert Systems with Applications: An International Journal
Complementary Kernel Density Estimation
Pattern Recognition Letters
Detonation Classification from Acoustic Signature with the Restricted Boltzmann Machine
Computational Intelligence
Editors Choice Article: I2VM: Incremental import vector machines
Image and Vision Computing
A geometric view of conjugate priors
IJCAI'11 Proceedings of the Twenty-Second international joint conference on Artificial Intelligence - Volume Volume Three
A hybrid generative/discriminative method for semi-supervised classification
Knowledge-Based Systems
Investigating Topic Models' Capabilities in Expression Microarray Data Classification
IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB)
A regularized linear classifier for effective text classification
ICONIP'12 Proceedings of the 19th international conference on Neural Information Processing - Volume Part II
Semi-supervised learning with density-ratio estimation
Machine Learning
Object class detection: A survey
ACM Computing Surveys (CSUR)
Pedestrian detection based on kernel discriminative sparse representation
Transactions on Edutainment IX
Effective automatic image annotation via integrated discriminative and generative models
Information Sciences: an International Journal
Co-trained generative and discriminative trackers with cascade particle filter
Computer Vision and Image Understanding
Enhancing K-Means using class labels
Intelligent Data Analysis
Hi-index | 0.00 |
When labelled training data is plentiful, discriminative techniques are widely used since they give excellent generalization performance. However, for large-scale applications such as object recognition, hand labelling of data is expensive, and there is much interest in semi-supervised techniques based on generative models in which the majority of the training data is unlabelled. Although the generalization performance of generative models can often be improved by 'training them discriminatively', they can then no longer make use of unlabelled data. In an attempt to gain the benefit of both generative and discriminative approaches, heuristic procedure have been proposed [2, 3] which interpolate between these two extremes by taking a convex combination of the generative and discriminative objective functions. In this paper we adopt a new perspective which says that there is only one correct way to train a given model, and that a 'discriminatively trained' generative model is fundamentally a new model [7]. From this viewpoint, generative and discriminative models correspond to specific choices for the prior over parameters. As well as giving a principled interpretation of 'discriminative training', this approach opens door to very general ways of interpolating between generative and discriminative extremes through alternative choices of prior. We illustrate this framework using both synthetic data and a practical example in the domain of multi-class object recognition. Our results show that, when the supply of labelled training data is limited, the optimum performance corresponds to a balance between the purely generative and the purely discriminative.