Corpus ID: 8070055

Deep Directed Generative Models with Energy-Based Probability Estimation

@article{Kim2016DeepDG,
  title={Deep Directed Generative Models with Energy-Based Probability Estimation},
  author={Taesup Kim and Yoshua Bengio},
  journal={ArXiv},
  year={2016},
  volume={abs/1606.03439}
}
  • Taesup Kim, Yoshua Bengio
  • Published 2016
  • Computer Science, Mathematics
  • ArXiv
  • Training energy-based probabilistic models is confronted with apparently intractable sums, whose Monte Carlo estimation requires sampling from the estimated probability distribution in the inner loop of training. This can be approximately achieved by Markov chain Monte Carlo methods, but may still face a formidable obstacle that is the difficulty of mixing between modes with sharp concentrations of probability. Whereas an MCMC process is usually derived from a given energy function based on… CONTINUE READING

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 79 CITATIONS

    Exponential Family Estimation via Adversarial Dynamics Embedding

    VIEW 2 EXCERPTS
    CITES METHODS & BACKGROUND

    Flow Contrastive Estimation of Energy-Based Models

    VIEW 3 EXCERPTS
    CITES METHODS & BACKGROUND

    Adversarial Variational Inference and Learning in Markov Random Fields

    VIEW 1 EXCERPT
    CITES BACKGROUND

    Energy-based Generative Adversarial Network

    VIEW 1 EXCERPT
    CITES BACKGROUND

    Generative Adversarial Networks as Variational Training of Energy Based Models

    VIEW 4 EXCERPTS
    CITES BACKGROUND
    HIGHLY INFLUENCED

    FILTER CITATIONS BY YEAR

    2016
    2020

    CITATION STATISTICS

    • 12 Highly Influenced Citations

    • Averaged 19 Citations per year from 2018 through 2020

    References

    Publications referenced by this paper.