Self-Correcting Self-Consuming Loops for Generative Model Training.
Nate GillmanMichael FreemanDaksh AggarwalChia-Hong HsuCalvin LuoYonglong TianChen SunPublished in: CoRR (2024)
Keyphrases
- generative model
- probabilistic model
- bayesian framework
- mixture model
- em algorithm
- prior knowledge
- object categories
- posterior probability
- conditional random fields
- discriminative learning
- latent dirichlet allocation
- semi supervised
- topic models
- discriminative models
- training examples
- markov chain monte carlo
- labelled data
- generative and discriminative models
- learning process
- training set
- data sets
- text classification
- text mining
- learned models
- multiscale
- primal sketch
- training data