Corpus ID: 3199842

Variational Auto-encoded Deep Gaussian Processes

@article{Dai2016VariationalAD,
  title={Variational Auto-encoded Deep Gaussian Processes},
  author={Z. Dai and A. Damianou and J. Gonz{\'a}lez and N. Lawrence},
  journal={CoRR},
  year={2016},
  volume={abs/1511.06455}
}
  • Z. Dai, A. Damianou, +1 author N. Lawrence
  • Published 2016
  • Computer Science, Mathematics
  • CoRR
  • We develop a scalable deep non-parametric generative model by augmenting deep Gaussian processes with a recognition model. Inference is performed in a novel scalable variational framework where the variational posterior distributions are reparametrized through a multilayer perceptron. The key aspect of this reformulation is that it prevents the proliferation of variational parameters which otherwise grow linearly in proportion to the sample size. We derive a new formulation of the variational… CONTINUE READING

    Paper Mentions

    Random Feature Expansions for Deep Gaussian Processes
    73
    Training Deep Gaussian Processes with Sampling
    4
    Practical learning of deep gaussian processes via random Fourier features
    14
    Stochastic Variational Deep Kernel Learning
    96
    Compositional uncertainty in deep Gaussian processes
    3
    Structured Variationally Auto-encoded Optimization
    10
    Deep Gaussian Processes with Importance-Weighted Variational Inference
    15
    Deep Gaussian Processes for Multi-fidelity Modeling
    16
    Neural Likelihoods for Multi-Output Gaussian Processes
    1

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 47 REFERENCES
    Auto-Encoding Variational Bayes
    8636
    Stochastic Backpropagation and Approximate Inference in Deep Generative Models
    2608
    Avoiding pathologies in very deep networks
    82
    Deep Generative Stochastic Networks Trainable by Backprop
    295
    Neural Variational Inference and Learning in Belief Networks
    509
    Generative Moment Matching Networks
    452
    A Deep and Tractable Density Estimator
    110
    Deep Gaussian Processes
    493
    Variational Dropout and the Local Reparameterization Trick
    527