Likelihood-free MCMC with Amortized Approximate Ratio Estimators

Hermans, Joeri; Begy, Volodimir; Louppe, Gilles

Download

Paper published in a book (Scientific congresses and symposiums)

Likelihood-free MCMC with Amortized Approximate Ratio Estimators

Hermans, Joeri; Begy, Volodimir; Louppe, Gilles

2020 • In Proceedings of the 37th International Conference on Machine Learning

Peer reviewed

Permalink
https://hdl.handle.net/2268/235737

Files (1)Send to Details Statistics Bibliography Similar publications

Files

Full Text

hermans20a.pdf

Publisher postprint (6.91 MB)

Download

All documents in ORBi are protected by a user license.

Send to

RIS BibTex APA Chicago Permalink X Linkedin

Details

Abstract :

[en] Posterior inference with an intractable likelihood is becoming an increasingly common task in scientific domains which rely on sophisticated computer simulations. Typically, these forward models do not admit tractable densities forcing practitioners to rely on approximations. This work introduces a novel approach to address the intractability of the likelihood and the marginal model. We achieve this by learning a flexible amortized estimator which approximates the likelihood-to-evidence ratio. We demonstrate that the learned ratio estimator can be embedded in \textsc{mcmc} samplers to approximate likelihood-ratios between consecutive states in the Markov chain, allowing us to draw samples from the intractable posterior. Techniques are presented to improve the numerical stability and to measure the quality of an approximation. The accuracy of our approach is demonstrated on a variety of benchmarks against well-established techniques. Scientific applications in physics show its applicability.

Disciplines :

Computer science

Author, co-author :

Hermans, Joeri ; Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Big Data

Begy, Volodimir

Louppe, Gilles ; Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Big Data

Language :

English

Title :

Likelihood-free MCMC with Amortized Approximate Ratio Estimators

Publication date :

July 2020

Event name :

37th International Conference on Machine Learning

Event date :

July 13-18, 2020

Audience :

International

Main work title :

Proceedings of the 37th International Conference on Machine Learning

Pages :

4239-4248

Peer reviewed :

Peer reviewed

Additional URL :

https://arxiv.org/abs/1903.04057

Available on ORBi :

since 22 May 2019

Statistics

Number of views

83 (6 by ULiège)

Number of downloads

420 (4 by ULiège)

More statistics

Scopus citations^®

Scopus citations^®
without self-citations

Bibliography

Azadi, S., Olsson, C., Darrell, T., Goodfellow, I., and Odena, A. Discriminator rejection sampling. arXiv preprint arXiv:1810.06758, 2018.
Baldi, P., Cranmer, K., Faucett, T., Sadowski, P., and Whiteson, D. Parameterized neural networks for high-energy physics. Eur. Phys. J. C, 76(5):235, April 2016. ISSN 1434-6044, 1434-6052. URL https://doi.org/10.1140/epjc/s10052-016-4099-4.
Beaumont, M. A., Zhang, W., and Balding, D. J. Approximate Bayesian computation in population genetics. Genetics, 162(4):2025-2035, 2002. URL http://www.genetics.org/content/162/4/2025.
Betancourt, M. A conceptual introduction to Hamiltonian Monte Carlo. arXiv preprint arXiv:1701.02434, 2017. URL https://arxiv.org/abs/1701.02434.
Boelts, J., Lueckmann, J.-M., Goncalves, P. J., Sprekeler, H., and Macke, J. H. Comparing neural simulations by neural density estimation. In 2019 Conference on Cognitive Computational Neuroscience, pp. 1289-1299.
Cognitive Computational Neuroscience, 2019. URL https://doi.org/10.32470/ccn.2019.1291-0.
Brehmer, J., Cranmer, K., Louppe, G., and Pavez, J. A guide to constraining effective field theories with machine learning. Phys. Rev. D, 98(5):052004, September 2018. ISSN 2470-0010, 2470-0029. URL https://doi.org/10.1103/physrevd.98.052004.
Brehmer, J., Louppe, G., Pavez, J., and Cranmer, K. Mining gold from implicit models to improve likelihood-free inference. Proc Natl Acad Sci USA, 117(10):5242-5249, February 2020. ISSN 0027-8424, 1091-6490. URL https://doi.org/10.1073/pnas.1915980117.
Clevert, D.-A., Unterthiner, T., and Hochreiter, S. Fast and accurate deep network learning by exponential linear units (elus). arXiv preprint arXiv:1511.07289, 2015.
Cranmer, K., Pavez, J., and Louppe, G. Approximating likelihood ratios with calibrated discriminative classifiers. arXiv preprint arXiv:1506.02169, 2015. URL https://arxiv.org/abs/1506.02169.
Crosby, O. PYTHIA, pythia, April 2020. URL https://doi.org/10.1002/9783527809080.cataz13886.
Dinev, T. and Gutmann, M. U. Dynamic likelihood-free inference via ratio estimation (dire). arXiv preprint arXiv:1810.09899, 2018.
Duane, S., Kennedy, A., Pendleton, B. J., and Roweth, D. Hybrid Monte Carlo. Phys. Lett. B, 195(2):216-222, September 1987. ISSN 0370-2693. URL https://doi.org/10.1016/0370-2693(87)91197-x.
Durkan, C., Murray, I., and Papamakarios, G. On Contrastive Learning for Likelihood-free Inference. arXiv e-prints, art. arXiv:2002.03712, February 2020.
Dutta, R., Corander, J., Kaski, S., and Gutmann, M. U. Likelihood-free inference by ratio estimation. arXiv preprint arXiv:1611.10242, 2016.
Fearnhead, P. and Prangle, D. Constructing summary statistics for approximate Bayesian computation: Semiautomatic approximate Bayesian computation. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 74(3):419-474, May 2012. ISSN 1369-7412. URL https://doi.org/10.1111/j.1467-9868.2011.01010.x.
Gershman, S. and Goodman, N. Amortized inference in probabilistic reasoning. In Proceedings of the Annual Meeting of the Cognitive Science Society, volume 36, 2014.
Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., and Bengio, Y. Generative adversarial nets. In Advances in neural information processing systems, pp. 2672-2680, 2014.
Greenberg, D., Nonnenmacher, M., and Macke, J. Automatic posterior transformation for likelihood-free inference. In Chaudhuri, K. and Salakhutdinov, R. (eds.), Proceedings of the 36th International Conference on Machine Learning, volume 97 of Proceedings of Machine Learning Research, pp. 2404-2414, Long Beach, California, USA, 2019. PMLR. URL http://proceedings.mlr.press/v97/greenberg19a.html.
Gretton, A., Borgwardt, K. M., Rasch, M. J., Schölkopf, B., and Smola, A. A kernel two-sample test. Journal of Machine Learning Research, 13(Mar):723-773, 2012.
Gutmann, M. U. and Corander, J. Bayesian optimization for likelihood-free inference of simulator-based statistical models. The Journal of Machine Learning Research, 17 (1):4256-4302, 2016. URL http://jmlr.org/papers/v17/15-017.html.
Gutmann, M. U., Dutta, R., Kaski, S., and Corander, J. Likelihood-free inference via classification. Stat Comput, 28(2):411-425, March 2017. ISSN 0960-3174, 1573-1375. URL https://doi.org/10.1007/s11222-017-9738-6.
Hastings, W. Monte Carlo sampling methods using Markov chains and their applications. Biometrika, 57(1):97-109, April 1970. ISSN 1464-3510, 0006-3444. URL https://doi.org/10.1093/biomet/57.1.97.
He, K., Zhang, X., Ren, S., and Sun, J. Deep residual learning for image recognition. In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770-778. IEEE, June 2016. URL https://doi.org/10.1109/cvpr.2016.90.
Hoffman, M. D., Blei, D. M., Wang, C., and Paisley, J. Stochastic variational inference. The Journal of Machine Learning Research, 14(1):1303-1347, 2013.
Ioffe, S. and Szegedy, C. Batch normalization: Accelerating deep network training by reducing internal covariate shift. arXiv preprint arXiv:1502.03167, 2015.
J. Neyman, E. P. IX. on the problem of the most efficient tests of statistical hypotheses. Phil. Trans. R. Soc. Lond. A, 231(694-706):289-337, February 1933. ISSN 0264-3952, 2053-9258. URL https://doi.org/10.1098/rsta.1933.0009.
Kingma, D. P. and Ba, J. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014.
Klambauer, G., Unterthiner, T., Mayr, A., and Hochreiter, S. Self-normalizing neural networks. In Advances in Neural Information Processing Systems, pp. 971-980, 2017.
Kormann, R., Schneider, P., and Bartelmann, M. Isothermal elliptical gravitational lens models. Astron. Astrophys., 284:285-299, 1994.
Lecun, Y., Bottou, L., Bengio, Y., and Haffner, P. Gradientbased learning applied to document recognition. Proc. IEEE, 86(11):2278-2324, 1998. ISSN 0018-9219. URL https://doi.org/10.1109/5.726791.
Lotka, A. J. Analytical note on certain rhythmic relations in organic systems. Proc Natl Acad Sci USA, 6(7):410-415, June 1920. ISSN 0027-8424, 1091-6490. URL https://doi.org/10.1073/pnas.6.7.410.
Louppe, G., Hermans, J., and Cranmer, K. Adversarial variational optimization of non-differentiable simulators. arXiv preprint arXiv:1707.07113, 2017. URL https://arxiv.org/abs/1707.07113.
Lueckmann, J.-M., Bassetto, G., Karaletsos, T., and Macke, J. H. Likelihood-free inference with emulator networks. arXiv preprint arXiv:1805.09294, 2018. URL https://arxiv.org/abs/1805.09294.
MacKay, D. J. Information theory, inference and learning algorithms. Cambridge university press, 2003.
Marin, J.-M., Pudlo, P., Robert, C. P., and Ryder, R. J. Approximate Bayesian computational methods. Stat Comput, 22(6):1167-1180, October 2011. ISSN 0960-3174, 1573-1375. URL https://doi.org/10.1007/s11222-011-9288-2.
Marjoram, P., Molitor, J., Plagnol, V., and Tavare, S. Markov chain Monte Carlo without likelihoods. Proceedings of the National Academy of Sciences, 100(26):15324-15328, December 2003. ISSN 0027-8424, 1091-6490. URL https://doi.org/10.1073/pnas.0306899100.
Meeds, E. andWelling, M. GPS-ABC: Gaussian process surrogate approximate Bayesian computation. arXiv preprint arXiv:1401.2838, 2014. URL https://arxiv.org/abs/1401.2838.
Metropolis, N., Rosenbluth, A. W., Rosenbluth, M. N., Teller, A. H., and Teller, E. Equation of state calculations by fast computing machines. The Journal of Chemical Physics, 21(6):1087-1092, June 1953. ISSN 0021-9606, 1089-7690. URL https://doi.org/10.1063/1.1699114.
Mohamed, S. and Lakshminarayanan, B. Learning in Implicit Generative Models. ArXiv e-prints, October 2016.
Neal, R. M. MCMC using Hamiltonian dynamics. Handbook of Markov Chain Monte Carlo, 2(11):2, 2011. URL https://arxiv.org/abs/1206.1901.
Neal, R. M. and Hinton, G. E. A view of the em algorithm that justifies incremental, sparse, and other variants. In Learning in Graphical Models, pp. 355-368. Springer Netherlands, 1998. URL https://doi.org/10.1007/978-94-011-5014-9_12.
Nightingale, J. W., Dye, S., and Massey, R. J. AutoLens: Automated modeling of a strong lens's light, mass, and source. Mon. Not. R. Astron. Soc., 478(4):4738-4784, May 2018. ISSN 0035-8711, 1365-2966. URL https://doi.org/10.1093/mnras/sty1264.
Ong, V. M., Nott, D. J., Tran, M.-N., Sisson, S. A., and Drovandi, C. C. Variational Bayes with synthetic likelihood. Stat Comput, 28(4):971-988, August 2017. ISSN 0960-3174, 1573-1375. URL https://doi.org/10.1007/s11222-017-9773-3.
Papamakarios, G. and Murray, I. Fast "-free inference of simulation models with Bayesian conditional density estimation. In Advances in Neural Information Processing Systems, pp. 1028-1036, 2016.
Papamakarios, G. and Murray, I. Sequential neural likelihood: Fast likelihood-free inference with autoregressive flows. arXiv preprint arXiv:1805.07226, 2018. URL https://arxiv.org/abs/1805.07226.
Paszke, A., Gross, S., Chintala, S., Chanan, G., Yang, E., DeVito, Z., Lin, Z., Desmaison, A., Antiga, L., and Lerer, A. Automatic differentiation in pytorch. 2017.
Pesah, A., Wehenkel, A., and Louppe, G. Recurrent machines for likelihood-free inference. arXiv preprint arXiv:1811.12932, 2018.
Pritchard, J., Seielstad, M., Perez-Lezaun, A., and Feldman, M. Population growth of human y chromosomes: A study of y chromosome microsatellites. Mol. Biol. Evol., 16(12):1791-1798, December 1999. ISSN 0737-4038, 1537-1719. URL https://doi.org/10.1093/oxfordjournals. molbev.a026091.
Ritchie, D., Horsfall, P., and Goodman, N. D. Deep amortized inference for probabilistic programs. arXiv preprint arXiv:1610.05735, 2016.
Salimans, T., Kingma, D., and Welling, M. Markov chain monte carlo and variational inference: Bridging the gap. In International Conference on Machine Learning, pp. 1218-1226, 2015.
Sjöstrand, T., Mrenna, S., and Skands, P. A brief introduction to PYTHIA 8.1. Comput. Phys. Commun., 178(11):852-867, June 2008. ISSN 0010-4655. URL https://doi.org/10.1016/j.cpc.2008.01.036.
Skands, P., Carrazza, S., and Rojo, J. Tuning PYTHIA 8.1: The monash 2013 tune. Eur. Phys. J. C, 74(8): 3024, August 2014. ISSN 1434-6044, 1434-6052. URL https://doi.org/10.1140/epjc/s10052-014-3024-y.
Sutton, R. S., McAllester, D. A., Singh, S. P., and Mansour, Y. Policy gradient methods for reinforcement learning with function approximation. In Advances in neural information processing systems, pp. 1057-1063, 2000.
Talts, S., Betancourt, M., Simpson, D., Vehtari, A., and Gelman, A. Validating Bayesian inference algorithms with simulation-based calibration. arXiv preprint arXiv:1804.06788, 2018.
Tavaré, S., Balding, D. J., Griffiths, R. C., and Donnelly, P. Inferring coalescence times from DNA sequence data. Genetics, 145(2):505-518, 1997. URL http://www.genetics.org/content/genetics/145/2/505.full.pdf.
Toni, T., Welch, D., Strelkowa, N., Ipsen, A., and Stumpf, M. P. Approximate Bayesian computation scheme for parameter inference and model selection in dynamical systems. J. R. Soc. Interface., 6(31):187-202, July 2008. ISSN 1742-5689, 1742-5662. URL https://doi.org/10.1098/rsif.2008.0172.
Tran, D., Ranganath, R., and Blei, D. Hierarchical implicit models and likelihood-free variational inference. In Advances in Neural Information Processing Systems, pp. 5523-5533, 2017.
Turner, R., Hung, J., Saatci, Y., and Yosinski, J. Metropolis-Hastings generative adversarial networks. arXiv preprint arXiv:1811.11357, 2018.
Uehara, M., Sato, I., Suzuki, M., Nakayama, K., and Matsuo, Y. Generative adversarial nets from a density ratio estimation perspective. arXiv preprint arXiv:1610.02920, 2016.
Wegmann, D., Leuenberger, C., and Excoffier, L. Efficient approximate Bayesian computation coupled with Markov chain Monte Carlo without likelihood. Genetics, 182 (4):1207-1218, June 2009. ISSN 0016-6731, 1943-2631. URL https://doi.org/10.1534/genetics.109.102509.
Williams, R. J. Simple statistical gradient-following algorithms for connectionist reinforcement learning. Mach Learn, 8(3-4):229-256, May 1992. ISSN 0885-6125, 1573-0565. URL https://doi.org/10.1007/bf00992696.
Wong, W., Jiang, B., Wu, T.-y., and Zheng, C. Learning summary statistic for approximate Bayesian computation via deep neural network. STAT SINICA, pp. 1595-1618, 2018. ISSN 1017-0405. URL https://doi.org/10.5705/ss.202015.0340