ACL RD-TEC 1.0 Summarization of P05-1043
Paper Title:
LEARNING STOCHASTIC OT GRAMMARS: A BAYESIAN APPROACH USING DATA AUGMENTATION AND GIBBS SAMPLING
LEARNING STOCHASTIC OT GRAMMARS: A BAYESIAN APPROACH USING DATA AUGMENTATION AND GIBBS SAMPLING
Primarily assigned technology terms:
- algorithm
- approximation
- bayesian approach
- computing
- conditional likelihood
- data augmentation
- density estimation
- em algorithm
- gibbs sampler
- gibbs sampling
- kernel
- kernel density estimation
- language acquisition
- learner
- learning
- learning algorithm
- learning algorithms
- learning method
- learning methods
- linguistic analysis
- matching
- maximum entropy
- maximum entropy model
- maximum-likelihood
- metropolis algorithm
- modeling
- monte-carlo method
- morphology
- optimization
- ranking
- sampling
- search
- sound learning
- tuning
Other assigned terms:
- approach
- bayesian framework
- bias
- case
- conditional distribution
- conditional independence
- conjunct
- constraint interaction
- convergence
- data model
- data set
- distribution
- entropy
- estimation
- exponential distribution
- fact
- formalisms
- frequency counts
- grammar
- grammars
- heuristic
- heuristic rules
- histogram
- hypotheses
- hypothesis
- hypothesis space
- ilokano reduplication
- implementation
- inferences
- joint distribution
- learning problem
- lexical items
- lexicon
- likelihood
- likelihood function
- likelihood ratio
- linguistic
- linguistic data
- linguistic theories
- linguistic theory
- linguistic variation
- linguistics
- linguistics literature
- linguists
- markov chain
- method
- noise
- normal distribution
- optimality theory
- parameter space
- posterior
- posterior distribution
- prior probability
- probabilities
- probability
- relation
- running time
- statistics
- suffixes
- surface form
- syntax
- target grammar
- term
- theories
- theory
- word
- words
