Geometry Score: A Method For Comparing Generative Adversarial Networks

@inproceedings{Khrulkov2018GeometrySA,
  title={Geometry Score: A Method For Comparing Generative Adversarial Networks},
  author={Valentin Khrulkov and Ivan V. Oseledets},
  booktitle={ICML},
  year={2018}
}
One of the biggest challenges in the research of generative adversarial networks (GANs) is assessing the quality of generated samples and detecting various levels of mode collapse. In this work, we construct a novel measure of performance of a GAN by comparing geometrical properties of the underlying data manifold and the generated one, which provides both qualitative and quantitative means for evaluation. Our algorithm can be applied to datasets of an arbitrary nature and is not limited to… CONTINUE READING
Recent Discussions
This paper has been referenced on Twitter 255 times over the past 90 days. VIEW TWEETS
2 Citations
24 References
Similar Papers

Citations

Publications citing this paper.
Showing 1-2 of 2 extracted citations

References

Publications referenced by this paper.
Showing 1-10 of 24 references

Similar Papers

Loading similar papers…