Abstract
Black-box optimization is primarily important for many computationally intensive applications, including reinforcement learning (RL), robot control, etc. This paper presents a novel theoretical framework for black-box optimization, in which our method performs stochastic updates with an implicit natural gradient of an exponential-family distribution. Theoretically, we prove the convergence rate of our framework with full matrix update for convex functions under Gaussian distribution. Our methods are very simple and contain fewer hyper-parameters than CMA-ES [12]. Empirically, our method with full matrix update achieves competitive performance compared with one of the state-of-the-art methods CMA-ES on benchmark test problems. Moreover, our methods can achieve high optimization precision on some challenging test functions (e.g., \(l_1\)-norm ellipsoid test problem and Levy test problem), while methods with explicit natural gradient, i.e., IGO [21] with full matrix update can not. This shows the efficiency of our methods.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Akimoto, Y., Nagata, Y., Ono, I., Kobayashi, S.: Bidirectional relation between CMA evolution strategies and natural evolution strategies. In: Schaefer, R., Cotta, C., Kołodziej, J., Rudolph, G. (eds.) PPSN 2010. LNCS, vol. 6238, pp. 154–163. Springer, Heidelberg (2010). https://doi.org/10.1007/978-3-642-15844-5_16
Amari, S.I.: Natural gradient works efficiently in learning. Neural Comput. 10(2), 251–276 (1998)
Amari, S.: Information Geometry and Its Applications. AMS, vol. 194. Springer, Tokyo (2016). https://doi.org/10.1007/978-4-431-55978-8
Azoury, K.S., Warmuth, M.K.: Relative loss bounds for on-line density estimation with the exponential family of distributions. Mach. Learn. 43(3), 211–246 (2001)
Back, T., Hoffmeister, F., Schwefel, H.P.: A survey of evolution strategies. In: Proceedings of the Fourth International Conference on Genetic Algorithms, vol. 2. Morgan Kaufmann Publishers, San Mateo (1991)
Balasubramanian, K., Ghadimi, S.: Zeroth-order (non)-convex stochastic optimization via conditional gradient and gradient updates. In: Advances in Neural Information Processing Systems, pp. 3455–3464 (2018)
Barsce, J.C., Palombarini, J.A., Martínez, E.C.: Towards autonomous reinforcement learning: automatic setting of hyper-parameters using Bayesian optimization. In: Computer Conference (CLEI), 2017 XLIII Latin American, pp. 1–9. IEEE (2017)
Bull, A.D.: Convergence rates of efficient global optimization algorithms. J. Mach. Learn. Res. (JMLR) 12, 2879–2904 (2011)
Choromanski, K., Pacchiano, A., Parker-Holder, J., Tang, Y.: From complexity to simplicity: adaptive ES-active subspaces for blackbox optimization. arXiv:1903.04268 (2019)
Choromanski, K., Rowland, M., Sindhwani, V., Turner, R.E., Weller, A.: Structured evolution with compact architectures for scalable policy optimization. In: ICML, pp. 969–977 (2018)
Domke, J.: Provable smoothness guarantees for black-box variational inference. arXiv preprint arXiv:1901.08431 (2019)
Hansen, N.: The CMA evolution strategy: a comparing review. In: Lozano, J.A., Larrañaga, P., Inza, I., Bengoetxea, E. (eds.) Towards a New Evolutionary Computation. STUDFUZZ, vol. 192, pp. 75–102. Springer, Heidelberg (2006). https://doi.org/10.1007/3-540-32494-1_4
Khan, M.E., Lin, W.: Conjugate-computation variational inference: converting variational inference in non-conjugate models to inferences in conjugate models. arXiv preprint arXiv:1703.04265 (2017)
Khan, M.E., Nielsen, D.: Fast yet simple natural-gradient descent for variational inference in complex models. In: 2018 International Symposium on Information Theory and Its Applications (ISITA), pp. 31–35. IEEE (2018)
Khan, M.E., Nielsen, D., Tangkaratt, V., Lin, W., Gal, Y., Srivastava, A.: Fast and scalable Bayesian deep learning by weight-perturbation in adam. In: ICML (2018)
Liu, G., et al.: Trust region evolution strategies. In: AAAI (2019)
Lizotte, D.J., Wang, T., Bowling, M.H., Schuurmans, D.: Automatic gait optimization with Gaussian process regression. In: IJCAI, vol. 7, pp. 944–949 (2007)
Lyu, Y.: Spherical structured feature maps for kernel approximation. In: Proceedings of the 34th International Conference on Machine Learning (ICML), pp. 2256–2264 (2017)
Lyu, Y., Yuan, Y., Tsang, I.W.: Efficient batch black-box optimization with deterministic regret bounds. arXiv preprint arXiv:1905.10041 (2019)
Negoescu, D.M., Frazier, P.I., Powell, W.B.: The knowledge-gradient algorithm for sequencing experiments in drug discovery. INFORMS J. Comput. 23(3), 346–363 (2011)
Ollivier, Y., Arnold, L., Auger, A., Hansen, N.: Information-geometric optimization algorithms: a unifying picture via invariance principles. J. Mach. Learn. Res. (JMLR) 18(1), 564–628 (2017)
Raskutti, G., Mukherjee, S.: The information geometry of mirror descent. IEEE Trans. Inf. Theory 61(3), 1451–1457 (2015)
Salimans, T., Ho, J., Chen, X., Sidor, S., Sutskever, I.: Evolution strategies as a scalable alternative to reinforcement learning. arXiv preprint arXiv:1703.03864 (2017)
Snoek, J., Larochelle, H., Adams, R.P.: Practical Bayesian optimization of machine learning algorithms. In: NeurIPS, pp. 2951–2959 (2012)
Srinivas, M., Patnaik, L.M.: Genetic algorithms: a survey. Computer 27(6), 17–26 (1994)
Srinivas, N., Krause, A., Kakade, S.M., Seeger, M.: Gaussian process optimization in the bandit setting: no regret and experimental design. In: ICML (2010)
Wang, G.G., Shan, S.: Review of metamodeling techniques in support of engineering design optimization. J. Mech. Des. 129(4), 370–380 (2007)
Wierstra, D., Schaul, T., Glasmachers, T., Sun, Y., Peters, J., Schmidhuber, J.: Natural evolution strategies. J. Mach. Learn. Res. (JMLR) 15(1), 949–980 (2014)
Xu, Z.: Deterministic sampling of sparse trigonometric polynomials. J. Complex. 27(2), 133–140 (2011)
Acknowledgement
We would like to thank all anonymous reviewers and the area chair for their valuable comments and suggestions. Yueming Lyu was supported by UTS President Scholarship. Ivor Tsang was supported by the Australian Research Council Grant (DP180100106 and DP200101328).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Lyu, Y., Tsang, I.W. (2021). Black-Box Optimizer with Stochastic Implicit Natural Gradient. In: Oliver, N., Pérez-Cruz, F., Kramer, S., Read, J., Lozano, J.A. (eds) Machine Learning and Knowledge Discovery in Databases. Research Track. ECML PKDD 2021. Lecture Notes in Computer Science(), vol 12977. Springer, Cham. https://doi.org/10.1007/978-3-030-86523-8_14
Download citation
DOI: https://doi.org/10.1007/978-3-030-86523-8_14
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-86522-1
Online ISBN: 978-3-030-86523-8
eBook Packages: Computer ScienceComputer Science (R0)
