Corpus ID: 70350020

Neural Empirical Bayes

@article{Saremi2019NeuralEB,
  title={Neural Empirical Bayes},
  author={S. Saremi and A. Hyv{\"a}rinen},
  journal={J. Mach. Learn. Res.},
  year={2019},
  volume={20},
  pages={181:1-181:23}
}
  • S. Saremi, A. Hyvärinen
  • Published 2019
  • Mathematics, Computer Science
  • J. Mach. Learn. Res.
  • We formulate a novel framework that unifies kernel density estimation and empirical Bayes, where we address a broad set of problems in unsupervised learning with a geometric interpretation rooted in the concentration of measure phenomenon. We start by energy estimation based on a denoising objective which recovers the original/clean data X from its measured/noisy version Y with empirical Bayes least squares estimator. The setup is rooted in kernel density estimation, but the log-pdf in Y is… CONTINUE READING
    Unnormalized Variational Bayes
    Nonparametric Score Estimators
    1
    Learning and Inference in Imaginary Noise Models
    1
    Solving Linear Inverse Problems Using the Prior Implicit in a Denoiser
    1
    AR-DAE: Towards Unbiased Neural Entropy Gradient Estimation
    1
    Homotopic Gradients of Generative Density Priors for MR Image Reconstruction
    Gaussian Gated Linear Networks

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 46 REFERENCES
    Deep Energy Estimator Networks
    29
    GSNs : Generative Stochastic Networks
    28
    Least Squares Estimation Without Priors or Supervision
    38
    Estimation of Non-Normalized Statistical Models by Score Matching
    380
    Adam: A Method for Stochastic Optimization
    47824
    Generalized Denoising Auto-Encoders as Generative Models
    342
    Gaussian Mean-Shift Is an EM Algorithm
    163
    Robust Exponential Memory in Hopfield Networks
    5
    Representation Learning: A Review and New Perspectives
    6108