Papers
Topics
Authors
Recent
Assistant
AI Research Assistant
Well-researched responses based on relevant abstracts and paper content.
Custom Instructions Pro
Preferences or requirements that you'd like Emergent Mind to consider when generating responses.
Gemini 2.5 Flash
Gemini 2.5 Flash 137 tok/s
Gemini 2.5 Pro 45 tok/s Pro
GPT-5 Medium 26 tok/s Pro
GPT-5 High 24 tok/s Pro
GPT-4o 116 tok/s Pro
Kimi K2 207 tok/s Pro
GPT OSS 120B 430 tok/s Pro
Claude Sonnet 4.5 37 tok/s Pro
2000 character limit reached

On Explicit Curvature Regularization in Deep Generative Models (2309.10237v1)

Published 19 Sep 2023 in cs.AI and cs.LG

Abstract: We propose a family of curvature-based regularization terms for deep generative model learning. Explicit coordinate-invariant formulas for both intrinsic and extrinsic curvature measures are derived for the case of arbitrary data manifolds embedded in higher-dimensional Euclidean space. Because computing the curvature is a highly computation-intensive process involving the evaluation of second-order derivatives, efficient formulas are derived for approximately evaluating intrinsic and extrinsic curvatures. Comparative studies are conducted that compare the relative efficacy of intrinsic versus extrinsic curvature-based regularization measures, as well as performance comparisons against existing autoencoder training methods. Experiments involving noisy motion capture data confirm that curvature-based methods outperform existing autoencoder regularization methods, with intrinsic curvature measures slightly more effective than extrinsic curvature measures.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (33)
  1. What regularized auto-encoders learn from the data-generating distribution. The Journal of Machine Learning Research, 15(1):3563–3593, 2014.
  2. A grassmann manifold handbook: Basic geometry and computational aspects. arXiv preprint arXiv:2011.13699, 2020.
  3. Besse, A. L. Einstein manifolds. Springer Science & Business Media, 2007.
  4. Optimizing the latent space of generative networks. arXiv preprint arXiv:1707.05776, 2017.
  5. Flows for simultaneous manifold learning and density estimation. In Advances in neural information processing systems, 2020.
  6. Learning flat latent manifolds with vaes. arXiv preprint arXiv:2002.04881, 2020.
  7. Do Carmo, M. P. Riemannian geometry, volume 6. Springer, 1992.
  8. Do Carmo, M. P. Differential geometry of curves and surfaces: revised and updated second edition. Courier Dover Publications, 2016.
  9. A report on harmonic maps. Bulletin of the London mathematical society, 10(1):1–68, 1978.
  10. Another report on harmonic maps. Bulletin of the London Mathematical Society, 20(5):385–524, 1988.
  11. Einstein, A. The general theory of relativity. In The Meaning of Relativity, pp.  54–75. Springer, 1922.
  12. Fecko, M. Differential geometry and Lie groups for physicists. Cambridge university press, 2006.
  13. Riemannian geometry for the statistical analysis of diffusion tensor data. Signal Processing, 87(2):250–262, 2007.
  14. Generative adversarial nets. In Advances in neural information processing systems, pp. 2672–2680, 2014.
  15. Hutchinson, M. F. A stochastic estimator of the trace of the influence matrix for laplacian smoothing splines. Communications in Statistics-Simulation and Computation, 18(3):1059–1076, 1989.
  16. Jang, C. Riemannian Distortion Measures for Non-Euclidean Data. PhD thesis, Seoul National University Graduate School, 2019.
  17. A riemannian geometric framework for manifold learning of non-euclidean data. Advances in Data Analysis and Classification, pp.  1–27, 2020.
  18. Geometrically regularized autoencoders for non-euclidean data. In The Eleventh International Conference on Learning Representations, 2022.
  19. Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114, 2013.
  20. Kramer, M. A. Nonlinear principal component analysis using autoassociative neural networks. AIChE journal, 37(2):233–243, 1991.
  21. Kühnel, W. Differential geometry, volume 77. American Mathematical Soc., 2015.
  22. Deep learning. nature, 521(7553):436–444, 2015.
  23. Lee, Y. Geometric Methods for Manifold Representation Learning. PhD thesis, Seoul National University Graduate School, 2023.
  24. Neighborhood reconstructing autoencoders. Advances in Neural Information Processing Systems, 34:536–546, 2021.
  25. A statistical manifold framework for point cloud data. In International Conference on Machine Learning, pp. 12378–12402. PMLR, 2022a.
  26. Regularized autoencoders for isometric representation learning. In International Conference on Learning Representations, 2022b. URL https://openreview.net/forum?id=mQxt8l7JL04.
  27. Ricci-Curbastro, G. Direzioni et invarianti principali in una varieta qualunque. Atti del Real Inst. Venezio, 63:1233–1239, 1904.
  28. Higher order contractive auto-encoder. In Joint European conference on machine learning and knowledge discovery in databases, pp.  645–660. Springer, 2011a.
  29. Contractive auto-encoders: Explicit invariance during feature extraction. In Icml, 2011b.
  30. Schutz, B. A first course in general relativity. Cambridge university press, 2022.
  31. Ntu rgb+ d: A large scale dataset for 3d human activity analysis. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp.  1010–1019, 2016.
  32. Stacked denoising autoencoders: Learning useful representations in a deep network with a local denoising criterion. Journal of machine learning research, 11(12), 2010.
  33. Wong, Y.-C. Differential geometry of grassmann manifolds. Proceedings of the National Academy of Sciences, 57(3):589–594, 1967.
Citations (11)

Summary

We haven't generated a summary for this paper yet.

Dice Question Streamline Icon: https://streamlinehq.com

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Lightbulb Streamline Icon: https://streamlinehq.com

Continue Learning

We haven't generated follow-up questions for this paper yet.

List To Do Tasks Checklist Streamline Icon: https://streamlinehq.com

Collections

Sign up for free to add this paper to one or more collections.