{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T09:27:00Z","timestamp":1791278820782,"version":"4.1.0"},"reference-count":23,"publisher":"MDPI AG","issue":"1","license":[{"start":{"date-parts":[[2015,1,13]],"date-time":"2015-01-13T00:00:00Z","timestamp":1421107200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Entropy"],"abstract":"<jats:p>Information geometric optimization (IGO) is a general framework for stochastic optimization problems aiming at limiting the influence of arbitrary parametrization choices: the initial problem is transformed into the optimization of a smooth function on a Riemannian manifold, defining a parametrization-invariant first order differential equation and, thus, yielding an approximately parametrization-invariant algorithm (up to second order in the step size). We define the geodesic IGO update, a fully parametrization-invariant algorithm using the Riemannian structure, and we compute it for the manifold of Gaussians, thanks to Noether\u2019s theorem. However, in similar algorithms, such as CMA-ES (Covariance Matrix Adaptation - Evolution Strategy) and xNES (exponential Natural Evolution Strategy), the time steps for the mean and the covariance are decoupled. We suggest two ways of doing so: twisted geodesic IGO (GIGO) and blockwise GIGO. Finally, we show that while the xNES algorithm is not GIGO, it is an instance of blockwise GIGO applied to the mean and covariance matrix separately. Therefore, xNES has an almost parametrization-invariant description.<\/jats:p>","DOI":"10.3390\/e17010304","type":"journal-article","created":{"date-parts":[[2015,1,13]],"date-time":"2015-01-13T11:28:59Z","timestamp":1421148539000},"page":"304-345","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":9,"title":["Black-Box Optimization Using Geodesics in Statistical Manifolds"],"prefix":"10.3390","volume":"17","author":[{"given":"J\u00e9r\u00e9my","family":"Bensadon","sequence":"first","affiliation":[{"name":"Laboratoire de Recherche en Informatique, Universit\u00e9 Paris-Sud, 91400 Orsay, France"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2015,1,13]]},"reference":[{"key":"ref_1","unstructured":"Ollivier, Y., Arnold, L., Auger, A., and Hansen, N. (2011). Information-geometric optimization algorithms: A unifying picture via invariance principles, arXiv, 1106.3708."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Amari, S.-I., and Nagaoka, H. (2007). Methods of Information Geometry (Translations of Mathematical Monographs), American Mathematical Society.","DOI":"10.1090\/mmono\/191"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"4260","DOI":"10.3390\/e16084260","article-title":"Combinatorial optimization with information geometry: The Newton method","volume":"16","author":"Pistone","year":"2014","journal-title":"Entropy"},{"key":"ref_4","unstructured":"Eriksen, P. (1986). Geodesics Connected with the Fisher Metric on the Multivariate Normal Manifold, Institute of Electronic Systems, Aalborg University."},{"key":"ref_5","first-page":"119","article-title":"An Explicit Solution of Information Geodesic Equations for the Multivariate Normal Model","volume":"9","author":"Calvo","year":"1991","journal-title":"Stat. Decis."},{"key":"ref_6","first-page":"125","article-title":"Remarks on geodesics for multivariate normal models","volume":"3","author":"Imai","year":"2011","journal-title":"J. Math-for-Industry"},{"key":"ref_7","first-page":"211","article-title":"A Riemannian geometry of the multivariate normal model","volume":"11","author":"Skovgaard","year":"1981","journal-title":"Scand. J. Stat."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"118","DOI":"10.1109\/TASSP.1986.1164786","article-title":"Computation of the Exact Information Matrix of Gaussian Time Series with Stationary Random Components","volume":"34","author":"Porat","year":"1986","journal-title":"IEEE Trans. Acoust. Speech Signal Process"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Baluja, S., and Caruana, R. (1995). Removing the Genetics from the Standard Genetic Algorithm, Morgan Kaufmann Publishers.","DOI":"10.1016\/B978-1-55860-377-6.50014-1"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Malag\u00f2, L., Matteucci, M., and Pistone, G. (2011, January 5\u20139). Towards the geometry of estimation of distribution algorithms based on the exponential family, Schwarzenberg, Austria.","DOI":"10.1145\/1967654.1967675"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"77","DOI":"10.1023\/B:NACO.0000023416.59689.4e","article-title":"Learning probability distributions in continuous evolutionary algorithms\u2014A comparative review","volume":"3","author":"Kern","year":"2003","journal-title":"Nat. Comput."},{"key":"ref_12","first-page":"949","article-title":"Natural evolution strategies","volume":"15","author":"Wierstra","year":"2014","journal-title":"J. Mach. Learn. Res."},{"key":"ref_13","unstructured":"Huang, W. (2013). Optimization Algorithms on Riemannian Manifolds with Applications. [Ph.D. Thesis, Florida State University]."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Absil, P.A., Mahony, R., and Sepulchre, R. (2008). Optimization Algorithms on Matrix Manifolds, Princeton University Press.","DOI":"10.1515\/9781400830244"},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Arnold, V., Vogtmann, K., and Weinstein, A. (1989). Mathematical Methods of Classical Mechanics (Graduate Texts in Mathematics), Springer.","DOI":"10.1007\/978-1-4757-2063-1"},{"key":"ref_16","unstructured":"Bourguignon, J. (2007). Calcul variationnel, Ecole Polytechnique. in French."},{"key":"ref_17","unstructured":"Jost, J., and Li-Jost, X. (1998). Calculus of Variations (Cambridge Studies in Advanced Mathematics), Cambridge University Press."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Gallot, S., Hulin, D., and LaFontaine, J. (2004). Riemannian Geometry (Universitext), Springer. [3rd ed].","DOI":"10.1007\/978-3-642-18855-8"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Glasmachers, T., Schaul, T., Yi, S., Wierstra, D., and Schmidhuber, J. (2010, January 7\u201311). Exponential natural evolution strategies, Portland, OR, USA.","DOI":"10.1145\/1830483.1830557"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Schaefer, R., Cotta, C., Ko\u0142odziej, J., and Rudolph, G. (2010). Parallel Problem Solving from Nature, PPSN XI, Springer.","DOI":"10.1007\/978-3-642-15844-5"},{"key":"ref_21","unstructured":"Hansen, N. Available online: https:\/\/www.lri.fr\/\u223chansen\/cmatutorial.pdf."},{"key":"ref_22","unstructured":"Bensadon, J. Available online: https:\/\/www.lri.fr\/~bensadon\/."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Akimoto, Y., and Ollivier, Y. (2013, January 16\u201320). Objective improvement in information-geometric optimization, Adelaide, Australia.","DOI":"10.1145\/2460239.2460241"}],"container-title":["Entropy"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1099-4300\/17\/1\/304\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T20:41:35Z","timestamp":1760215295000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1099-4300\/17\/1\/304"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2015,1,13]]},"references-count":23,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2015,1]]}},"alternative-id":["e17010304"],"URL":"https:\/\/doi.org\/10.3390\/e17010304","relation":{},"ISSN":["1099-4300"],"issn-type":[{"value":"1099-4300","type":"electronic"}],"subject":[],"published":{"date-parts":[[2015,1,13]]}}}