Diversity vs. Recognizability: Human-like generalization in one-shot generative models.
Victor BoutinLakshya SinghalXavier ThomasThomas SerrePublished in: CoRR (2022)
Keyphrases
- generative model
- probabilistic model
- representational power
- mixture model
- prior knowledge
- discriminative models
- hierarchical hidden markov models
- em algorithm
- semi supervised
- discriminative learning
- expectation maximization
- generative and discriminative models
- object categories
- hidden variables
- maximum entropy principle
- conditional random fields
- similarity measure
- topic models
- semi supervised learning
- image processing