Corpus ID: 59606242

Latent Space Cartography: Generalised Metric-Inspired Measures and Measure-Based Transformations for Generative Models

@article{Frenzel2019LatentSC,
  title={Latent Space Cartography: Generalised Metric-Inspired Measures and Measure-Based Transformations for Generative Models},
  author={Max F. Frenzel and Bogdan Teleaga and A. Ushio},
  journal={ArXiv},
  year={2019},
  volume={abs/1902.02113}
}
  • Max F. Frenzel, Bogdan Teleaga, A. Ushio
  • Published 2019
  • Mathematics, Computer Science
  • ArXiv
  • Deep generative models are universal tools for learning data distributions on high dimensional data spaces via a mapping to lower dimensional latent spaces. We provide a study of latent space geometries and extend and build upon previous results on Riemannian metrics. We show how a class of heuristic measures gives more flexibility in finding meaningful, problem-specific distances, and how it can be applied to diverse generator types such as autoregressive generators commonly used in e.g… CONTINUE READING
    The Input/Output Complexity
    • JEFFREY SCOTT VlllER
    • 1999
    Intrinsic Point Cloud Interpolation via Dual Latent Space Navigation

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 21 REFERENCES
    Differential modulation of brain immediate early genes by intraperitoneal LiCl
    72
    Self-regulation and depletion of limited resources: does self-control resemble a muscle?
    3317
    Normalization of Tumor Vasculature: An Emerging Concept in Antiangiogenic Therapy
    4285
    From The Cover: Diffusion-based method for producing density-equalizing maps.
    481
    Feature-Based Metrics for Exploring the Latent Space of Generative Models
    14
    Latent Space Oddity: on the Curvature of Deep Generative Models
    67