Maximum Margin Training of Gaussian HMMs for Handwriting Recognition.
Trinh Minh Tri DoThierry ArtièresPublished in: ICDAR (2009)
Keyphrases
- maximum margin
- handwriting recognition
- structured prediction
- speech recognition
- hidden markov models
- support vector
- maximum likelihood
- markov networks
- learning algorithm
- character recognition
- hyperplane
- training set
- word recognition
- pattern classification
- support vector machine
- handwritten text
- word segmentation
- supervised learning
- neural network
- conditional random fields
- generalization error
- training samples
- expectation maximization
- training examples
- multi task
- multiple kernel learning
- graphical models
- probabilistic model