Topic Modeling with Wasserstein Autoencoders

@inproceedings{Nan2019TopicMW,
  title={Topic Modeling with Wasserstein Autoencoders},
  author={Feng Nan and Ran Ding and Ramesh Nallapati and B. Xiang},
  booktitle={ACL},
  year={2019}
}
  • Feng Nan, Ran Ding, +1 author B. Xiang
  • Published in ACL 2019
  • Computer Science
  • We propose a novel neural topic model in the Wasserstein autoencoders (WAE) framework. [...] Key Method We exploit the structure of the latent space and apply a suitable kernel in minimizing the Maximum Mean Discrepancy (MMD) to perform distribution matching. We discover that MMD performs much better than the Generative Adversarial Network (GAN) in matching high dimensional Dirichlet distribution. We further discover that incorporating randomness in the encoder output during training leads to significantly more…Expand Abstract

    Figures, Tables, and Topics from this paper.

    Neural Sinkhorn Topic Model
    Improving Neural Topic Models using Knowledge Distillation
    Neural Mixed Counting Models for Dispersed Topic Discovery
    Neural Topic Modeling by Incorporating Document Relationship Graph
    Multi-turn Response Selection using Dialogue Dependency Relations

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 110 REFERENCES
    Wasserstein Auto-Encoders
    • 332
    • Highly Influential
    • PDF
    Improved Training of Wasserstein GANs
    • 3,338
    • PDF
    Adversarial Autoencoders
    • 831
    • Highly Influential
    • PDF
    Adversarially Regularized Autoencoders
    • 136
    • PDF
    Generative Adversarial Nets
    • 17,848
    • PDF
    Semi-Amortized Variational Autoencoders
    • 107
    • PDF
    Neural Variational Inference for Text Processing
    • 324
    • Highly Influential
    • PDF
    Auto-Encoding Variational Bayes
    • 9,185
    • Highly Influential
    • PDF
    Autoencoding Variational Inference For Topic Models
    • 153
    • Highly Influential
    • PDF
    Lagging Inference Networks and Posterior Collapse in Variational Autoencoders
    • 91
    • PDF