Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
119 tokens/sec
GPT-4o
56 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
6 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

A VAE Approach to Sample Multivariate Extremes (2306.10987v1)

Published 19 Jun 2023 in stat.ML and cs.LG

Abstract: Generating accurate extremes from an observational data set is crucial when seeking to estimate risks associated with the occurrence of future extremes which could be larger than those already observed. Applications range from the occurrence of natural disasters to financial crashes. Generative approaches from the machine learning community do not apply to extreme samples without careful adaptation. Besides, asymptotic results from extreme value theory (EVT) give a theoretical framework to model multivariate extreme events, especially through the notion of multivariate regular variation. Bridging these two fields, this paper details a variational autoencoder (VAE) approach for sampling multivariate heavy-tailed distributions, i.e., distributions likely to have extremes of particularly large intensities. We illustrate the relevance of our approach on a synthetic data set and on a real data set of discharge measurements along the Danube river network. The latter shows the potential of our approach for flood risks' assessment. In addition to outperforming the standard VAE for the tested data sets, we also provide a comparison with a competing EVT-based generative approach. On the tested cases, our approach improves the learning of the dependency structure between extremes.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (58)
  1. Ev-gan: Simulation of extreme events with relu neural networks. Journal of Machine Learning Research, 23(150):1–39, 2022.
  2. Modeling river flows with heavy tails. Water Resources Research, 34(9):2271–2280, 1998.
  3. Wasserstein generative adversarial networks. In International conference on machine learning, pages 214–223. PMLR, 2017.
  4. Understanding deep neural networks with rectified linear units. arXiv preprint arXiv:1611.01491, 2016.
  5. Extremes on river networks. The Annals of Applied Statistics, 9(4):2023–2050, 2015.
  6. Residual life time at great age. The Annals of probability, 2(5):792–804, 1974.
  7. Regularly varying multivariate time series. Stochastic processes and their applications, 119(4):1055–1080, 2009.
  8. Exgan: Adversarial generation of extreme samples. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, pages 6750–6758, 2021.
  9. Regular variation. Number 27. Cambridge university press, 1989.
  10. Modeling and simulating spatial extremes by combining extreme value theory with generative adversarial networks. Environmental Data Science, 1, 2022.
  11. Financial risk and heavy tails. In Handbook of heavy tailed distributions in finance, pages 35–103. Elsevier, 2003.
  12. Leonard Breiman. On some limit theorems similar to the arc-sin law. Theory of Probability & Its Applications, 10(2):323–331, 1965.
  13. Extreme value theory can save your neck. ETHZ publication, 2004.
  14. New exponential bounds and approximations for the computation of error probability in fading channels. IEEE Transactions on Wireless Communications, 2(4):840–845, 2003.
  15. Concentration bounds for the empirical angular measure with statistical learning applications. arXiv preprint arXiv:2104.03966, 2021.
  16. An introduction to statistical modeling of extreme values, volume 208. Springer, 2001.
  17. Tail index estimation: Quantile driven threshold selection. Available at SSRN 2717478, 2016.
  18. Four theorems and a financial crisis. International Journal of Approximate Reasoning, 54(6):701–716, 2013.
  19. Sea and wind: multivariate extremes at work. Extremes, 1:7–45, 1998.
  20. Principal component analysis for multivariate extremes. Electronic Journal of Statistics, 15(1):908–943, 2021.
  21. Estimating extreme bivariate quantile regions. Extremes, 16(2):121–145, 2013.
  22. Paul Embrechts. Copulas: A personal view. Journal of Risk and Insurance, 76(3):639–650, 2009.
  23. Extreme value theory as a risk management tool. North American Actuarial Journal, 3(2):30–41, 1999.
  24. Robust bounds in multivariate extremes. 2017.
  25. Nonlinear 3d cosmic web simulation with heavy-tailed generative adversarial networks. Physical Review D, 102(10):103504, 2020.
  26. Implicit reparameterization gradients. Advances in neural information processing systems, 31, 2018.
  27. Pot: Python optimal transport. J. Mach. Learn. Res., 22(78):1–8, 2021.
  28. Applications of extreme value statistics in physics. Journal of Physics A: Mathematical and Theoretical, 48(18):183001, 2015.
  29. Generative adversarial networks. Communications of the ACM, 63(11):139–144, 2020.
  30. Ian Grooms. Analog ensemble data assimilation and a method for constructing analogs with variational autoencoders. Quarterly Journal of the Royal Meteorological Society, 147(734):139–149, 2021.
  31. Variable heavy tails in internet traffic. Performance Evaluation, 58(2-3):261–284, 2004.
  32. Pareto gan: Extending the representational power of gans to heavy-tailed distributions. In International Conference on Machine Learning, pages 4523–4532. PMLR, 2021.
  33. Tails of lipschitz triangular flows. In International Conference on Machine Learning, pages 4673–4681. PMLR, 2020.
  34. On binary classification in extreme regions. Advances in Neural Information Processing Systems, 31, 2018.
  35. Statistics of extremes in hydrology. Advances in water resources, 25(8-12):1287–1304, 2002.
  36. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014.
  37. Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114, 2013.
  38. Marginal tail-adaptive normalizing flows. In International Conference on Machine Learning, pages 12020–12048. PMLR, 2022.
  39. Sparse-gev: Sparse latent space model for multivariate extreme value time serie modeling. arXiv preprint arXiv:1206.4685, 2012.
  40. Causal mechanism of extreme river discharges in the upper danube basin network. Journal of the Royal Statistical Society: Series C (Applied Statistics), 69(4):741–764, 2020.
  41. Generating images with sparse representations. arXiv preprint arXiv:2103.03841, 2021.
  42. A non-parametric entropy-based approach to detect changes in climate extremes. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 76(5):861–884, 2014.
  43. Neural networks for extreme quantile regression with an application to forecasting of flood risk. arXiv preprint arXiv:2208.07590, 2022.
  44. Variational bayes. Elsevier, London, 2006.
  45. James Pickands III. Statistical inference using extreme order statistics. the Annals of Statistics, pages 119–131, 1975.
  46. Generating diverse high-fidelity images with vq-vae-2. Advances in neural information processing systems, 32, 2019.
  47. Sidney I Resnick. Heavy-tail phenomena: probabilistic and statistical modeling. Springer Science & Business Media, 2007.
  48. Variational inference with normalizing flows. In International conference on machine learning, pages 1530–1538. PMLR, 2015.
  49. Stochastic backpropagation and approximate inference in deep generative models. In International conference on machine learning, pages 1278–1286. PMLR, 2014.
  50. Network design for heavy rainfall analysis. Journal of Geophysical Research: Atmospheres, 118(23):13–075, 2013.
  51. Multivariate generalized pareto distributions. Bernoulli, 12(5):917–930, 2006.
  52. The extreme value machine. IEEE transactions on pattern analysis and machine intelligence, 40(3):762–768, 2017.
  53. Measuring and testing dependence by correlation of distances. The annals of statistics, 35(6):2769–2794, 2007.
  54. Flexible semiparametric generalized Pareto modeling of the entire range of rainfall amount. Environmetrics, 31(2):e2582, 2019.
  55. Threshold selection for multivariate heavy-tailed data. Extremes, 22(1):131–166, 2019.
  56. Xiaolei Xie. Analysis of Heavy-Tailed Time Series. PhD thesis, University of Copenhagen, Faculty of Science, Department of Mathematical …, 2017.
  57. Learning discourse-level diversity for neural dialog models using conditional variational autoencoders. arXiv preprint arXiv:1703.10960, 2017.
  58. Pluralistic image completion. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 1438–1447, 2019.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (3)
  1. Nicolas Lafon (1 paper)
  2. Philippe Naveau (27 papers)
  3. Ronan Fablet (53 papers)
Citations (2)
View on Semantic Scholar

Summary

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

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now