Divergence Frontiers for Generative Models: Sample Complexity, Quantization Effects, and Frontier Integrals.
Lang LiuKrishna PillutlaSean WelleckSewoong OhYejin ChoiZaïd HarchaouiPublished in: NeurIPS (2021)
Keyphrases
- generative model
- sample complexity
- theoretical analysis
- probabilistic model
- learning problems
- supervised learning
- lower bound
- learning algorithm
- active learning
- discriminative learning
- upper bound
- special case
- discriminative models
- prior knowledge
- generalization error
- semi supervised
- training examples
- expectation maximization
- em algorithm
- semi supervised learning
- generative and discriminative models
- hierarchical hidden markov models
- sample size
- conditional random fields
- computational complexity
- learning tasks
- maximum likelihood
- data mining
- data points
- learning process
- objective function
- reinforcement learning
- machine learning