Related Topics
Experts

1

2

3

4

5

6

7

8

9

10
Publications (14)

1Harnessing the CRF Complexity with DomainSpecific Constraints. The Case of Morphosyntactic Tagging of a Highly Inflected Language2012  Jakub Waszczukprobable label sequence given the sentence and assigning marginal probabilities to individual labels from restricted sets. The first...

2Modeling LatentDynamic in Shallow Parsing: A Latent Conditional Model with Imrpoved Inference(BMP). To estimate the label y j of token j, the marginal probabilities P(h j = ax,Θ) are computed for possible hidden states...

3Latent Mixture of Discriminative Experts for Multimodal Prediction Modelinge., P (λα) ∼ exp ( 1 2σ2 λα2 ) . Then the marginal probabilities Pα(yj = a  y, x, λ∗α), are computed using belief...

4Mining Largescale Comparable Corpora from ChineseEnglish News Collectionsspeech tagging for Chinese. The method based on the marginal probabilities detailed in (Luo and Huang, 2009) is adopted in ...

5Semisupervised Representation Learning for Domain Adaptation using Dynamic Dependency Networksforward and backward procedures used for HMMs. Then the marginal probabilities can be computed as P(Yt = y X , Z ,θ) = αt(y)βt(y)∑by...

6Exact Inference for Multilabel Classification using Sparse Graphical Models2008  Yusuke Miyao,Jun’ichi Tsujiiin order to compute argmax in Equation (1), or the marginal probabilities of cliques and labels, necessary for the parameter...

7Kvec: A New Approach for Aligning Parallel Texts1994  Pascale Fung,Kenneth Ward Churchpiece is simply: a prob(Vf, Vp)  a+b+c+d The marginal probabilities are: a+b prob(Vf )  a+b+c+d a+c prob(Vp)...

8A Separately PassiveAggressive Training Algorithm for Joint POS Tagging and Dependency Parsingprune out the implausible POS tags according to the marginal probabilities (see Section 4.1) and list the top three candidate...

9A Latent Discriminative Model for Compositional Entailment Relation Recognition using Natural Logicorder to update the parameters, we need to calculate marginal probabilities of the alignments. However, unlike sequential or...

10Large Scale Parallel Document Mining for Machine Translationp(t) (3) where the joint probabilities p(s, t) and marginal probabilities p(s), p(t) are taken to be the respective empirical...