Oops I Took A Gradient: Scalable Sampling for Discrete Distributions.
Will GrathwohlKevin SwerskyMilad HashemiDavid DuvenaudChris J. MaddisonPublished in: CoRR (2021)
Keyphrases
- efficient algorithms to compute
- random sampling
- band limited
- discrete geometry
- finite number
- highly skewed
- edge detection
- lightweight
- sampling strategy
- sampled data
- highly scalable
- random samples
- translation invariant
- kullback leibler divergence
- sampling algorithm
- probability distribution
- active learning
- multiscale
- learning algorithm
- sampling methods
- power law
- continuous data
- computational geometry
- gaussian distribution
- sample size
- heavy tailed
- random variables
- generative model