Principled Gradient-based Markov Chain Monte Carlo for Text Generation.
Li DuAfra AminiLucas Torroba HennigenXinyan Velocity YuJason EisnerHolden LeeRyan CotterellPublished in: CoRR (2023)
Keyphrases
- markov chain monte carlo
- text generation
- natural language generation
- parameter estimation
- markov chain
- generative model
- monte carlo
- natural language
- posterior distribution
- bayesian inference
- particle filter
- posterior probability
- approximate inference
- theorem prover
- markov chain monte carlo sampling
- particle filtering
- data association
- simulated annealing
- graphical models
- incremental learning
- object tracking
- probabilistic inference
- first order logic
- cross validation
- topic models
- em algorithm
- higher order
- active learning