• Publications
  • Influence
Masked Autoregressive Flow for Density Estimation
TLDR
We describe an approach for increasing the flexibility of an autoregressive model, based on modelling the random numbers that the model uses internally when generating data. Expand
  • 424
  • 91
  • PDF
MADE: Masked Autoencoder for Distribution Estimation
TLDR
We introduce a simple modification for autoencoder neural networks that yields powerful generative models that are faster than other autoregressive estimators. Expand
  • 372
  • 70
  • PDF
On the quantitative analysis of deep belief networks
TLDR
We show that Annealed Importance Sampling (AIS) can be used to efficiently estimate the partition function of an RBM, and we present a novel AIS scheme for comparing RBM's with different architectures. Expand
  • 418
  • 68
  • PDF
The Neural Autoregressive Distribution Estimator
TLDR
We describe a new approach for modeling the distribution of high-dimensional vectors of discrete variables. Expand
  • 403
  • 66
  • PDF
MCMC for Doubly-intractable Distributions
TLDR
This paper provides a generalization of Moller et al. (2004) and a new MCMC algorithm, which obtains better acceptance probabilities for the same amount of exact sampling, and removes the need to estimate model parameters before sampling begins. Expand
  • 321
  • 63
  • PDF
Elliptical slice sampling
TLDR
We present a new Markov chain Monte Carlo algorithm for performing inference in models with multivariate Gaussian priors. Expand
  • 327
  • 51
  • PDF
Evaluation methods for topic models
TLDR
A natural evaluation metric for statistical topic models is the probability of held-out documents given a trained model. Expand
  • 701
  • 49
  • PDF
Tractable nonparametric Bayesian inference in Poisson processes with Gaussian process intensities
TLDR
In this paper we present the first approach to Gaussian Cox processes in which it is possible to perform inference without introducing approximations or finitedimensional proxy distributions. Expand
  • 177
  • 47
  • PDF
Introduction to Gaussian Processes
  • 287
  • 25
  • PDF
Fast ε-free Inference of Simulation Models with Bayesian Conditional Density Estimation
TLDR
We propose a new approach to likelihood-free inference based on Bayesian conditional density estimation, which can be made as accurate as required. Expand
  • 104
  • 24
  • PDF