{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,21]],"date-time":"2026-11-21T03:19:47Z","timestamp":1795231187122,"version":"4.10.3"},"reference-count":32,"publisher":"MDPI AG","issue":"2","license":[{"start":{"date-parts":[[2022,1,27]],"date-time":"2022-01-27T00:00:00Z","timestamp":1643241600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100004837","name":"Ministerio de Ciencia e Innovaci\u00f3n","doi-asserted-by":"publisher","award":["PID2019-106426RB-C31"],"award-info":[{"award-number":["PID2019-106426RB-C31"]}],"id":[{"id":"10.13039\/501100004837","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Entropy"],"abstract":"<jats:p>In this paper, we propose a novel and generic family of multiple importance sampling estimators. We first revisit the celebrated balance heuristic estimator, a widely used Monte Carlo technique for the approximation of intractable integrals. Then, we establish a generalized framework for the combination of samples simulated from multiple proposals. Our approach is based on considering as free parameters both the sampling rates and the combination coefficients, which are the same in the balance heuristics estimator. Thus our novel framework contains the balance heuristic as a particular case. We study the optimal choice of the free parameters in such a way that the variance of the resulting estimator is minimized. A theoretical variance study shows the optimal solution is always better than the balance heuristic estimator (except in degenerate cases where both are the same). We also give sufficient conditions on the parameter values for the new generalized estimator to be better than the balance heuristic estimator, and one necessary and sufficient condition related to \u03c72 divergence. Using five numerical examples, we first show the gap in the efficiency of both new and classical balance heuristic estimators, for equal sampling and for several state of the art sampling rates. Then, for these five examples, we find the variances for some notable selection of parameters showing that, for the important case of equal count of samples, our new estimator with an optimal selection of parameters outperforms the classical balance heuristic. Finally, new heuristics are introduced that exploit the theoretical findings.<\/jats:p>","DOI":"10.3390\/e24020191","type":"journal-article","created":{"date-parts":[[2022,1,27]],"date-time":"2022-01-27T21:59:55Z","timestamp":1643320795000},"page":"191","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":13,"title":["Generalizing the Balance Heuristic Estimator in Multiple Importance Sampling"],"prefix":"10.3390","volume":"24","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2164-6858","authenticated-orcid":false,"given":"Mateu","family":"Sbert","sequence":"first","affiliation":[{"name":"Informatics and Applications Institute, Girona University, 17003 Girona, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"V\u00edctor","family":"Elvira","sequence":"additional","affiliation":[{"name":"School of Mathematics, University of Edinburgh, Edinburgh EH8 9YL, UK"},{"name":"The Alan Turing Institute, London NW1 2DB, UK"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2022,1,27]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Veach, E., and Guibas, L. (1995, January 6\u201311). Optimally Combining Sampling Techniques for Monte Carlo Rendering. Proceedings of the 22nd Annual Conference on Computer Graphics and Interactive Techniques, Los Angeles, CA, USA.","DOI":"10.1145\/218380.218498"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"129","DOI":"10.1214\/18-STS668","article-title":"Generalized Multiple Importance Sampling","volume":"34","author":"Elvira","year":"2019","journal-title":"Stat. Sci."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"135","DOI":"10.1080\/01621459.2000.10473909","article-title":"Safe and Effective Importance Sampling","volume":"95","author":"Owen","year":"2000","journal-title":"J. Am. Stat. Assoc."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"1757","DOI":"10.1109\/LSP.2015.2432078","article-title":"Efficient Multiple Importance Sampling Estimators","volume":"22","author":"Elvira","year":"2015","journal-title":"IEEE Signal Process. Lett."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Elvira, V., Martino, L., Luengo, D., and Bugallo, M.F. (2016, January 26\u201329). Multiple importance sampling with overlapping sets of proposals. Proceedings of the Statistical Signal Processing Workshop (SSP), Palma de Mallorca, Spain.","DOI":"10.1109\/SSP.2016.7551744"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"1474","DOI":"10.1109\/LSP.2016.2600678","article-title":"Heretical Multiple Importance Sampling","volume":"23","author":"Elvira","year":"2016","journal-title":"IEEE Signal Process. Lett."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"451","DOI":"10.1111\/cgf.13042","article-title":"Variance Analysis of Multi-sample and One-sample Multiple Importance Sampling","volume":"35","author":"Sbert","year":"2016","journal-title":"Comput. Graph. Forum"},{"key":"ref_8","unstructured":"Havran, V., and Sbert, M. (December, January 30). Optimal Combination of Techniques in Multiple Importance Sampling. Proceedings of the 13th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and Its Applications in Industry, VRCAI \u201914, Shenzhen, China."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1007\/s00371-017-1398-1","article-title":"Adaptive multiple importance sampling for general functions","volume":"33","author":"Sbert","year":"2017","journal-title":"Vis. Comput."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"843","DOI":"10.1007\/s00371-018-1522-x","article-title":"Multiple importance sampling characterization by weighted mean invariance","volume":"34","author":"Sbert","year":"2018","journal-title":"Vis. Comput."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"907","DOI":"10.1198\/106186004X12803","article-title":"Population Monte Carlo","volume":"13","author":"Guillin","year":"2004","journal-title":"J. Comput. Graph. Stat."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"798","DOI":"10.1111\/j.1467-9469.2011.00756.x","article-title":"Adaptive Multiple Importance Sampling","volume":"39","author":"Cornuet","year":"2012","journal-title":"Scand. J. Stat."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"599","DOI":"10.1007\/s11222-016-9642-5","article-title":"Layered adaptive importance sampling","volume":"27","author":"Martino","year":"2017","journal-title":"Stat. Comput."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"77","DOI":"10.1016\/j.sigpro.2016.07.012","article-title":"Improving Population Monte Carlo: Alternative Weighting and Resampling Schemes","volume":"131","author":"Elvira","year":"2017","journal-title":"Signal Process."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"60","DOI":"10.1109\/MSP.2017.2699226","article-title":"Adaptive Importance Sampling: The past, the present, and the future","volume":"34","author":"Bugallo","year":"2017","journal-title":"IEEE Signal Process. Mag."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"37","DOI":"10.1145\/3306346.3323009","article-title":"Optimal Multiple Importance Sampling","volume":"38","author":"Kondapaneni","year":"2019","journal-title":"ACM Trans. Graph."},{"key":"ref_17","unstructured":"Veach, E. (1997). Robust Monte Carlo Methods for Light Transport Simulation. [Ph.D. Thesis, Stanford University]."},{"key":"ref_18","first-page":"151","article-title":"MIS Compensation: Optimizing Sampling Techniques in Multiple Importance Sampling","volume":"38","year":"2019","journal-title":"ACM Trans. Graph."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"152","DOI":"10.1145\/3355089.3356515","article-title":"Variance-Aware Multiple Importance Sampling","volume":"38","author":"Grittmann","year":"2019","journal-title":"ACM Trans. Graph."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"136","DOI":"10.1145\/3386569.3392436","article-title":"Continuous Multiple Importance Sampling","volume":"39","author":"West","year":"2020","journal-title":"ACM Trans. Graph."},{"key":"ref_21","unstructured":"Evans, D.C., and Athay, R.J. (1986, January 18\u201322). The Rendering Equation. Proceedings of the Computer Graphics (SIGGRAPH \u201986 Proceedings), Dallas, TX, USA."},{"key":"ref_22","unstructured":"Robert, C.P. (2007). The Bayesian Choice: From Decision-Theoretic Foundations to Computational Implementation, Springer."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"474","DOI":"10.1109\/TSP.2020.3045526","article-title":"Importance Gaussian Quadrature","volume":"69","author":"Elvira","year":"2020","journal-title":"IEEE Trans. Signal Process."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Elvira, V., and Chouzenoux, E. (2019, January 12\u201317). Langevin-based strategy for efficient proposal adaptation in population Monte Carlo. Proceedings of the ICASSP 2019\u20142019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Brighton, UK.","DOI":"10.1109\/ICASSP.2019.8682284"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Rubinstein, R., and Kroese, D. (2008). Simulation and the Monte Carlo Method, Wiley.","DOI":"10.1002\/9780470230381"},{"key":"ref_26","first-page":"28","article-title":"Stochastic Orders on Two-Dimensional Space: Application to Cross Entropy","volume":"Volume 12256","author":"Torra","year":"2020","journal-title":"Proceedings of the Modeling Decisions for Artificial Intelligence\u201417th International Conference, MDAI 2020"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"292","DOI":"10.1186\/s13660-016-1233-7","article-title":"A necessary and sufficient condition for the inequality of generalized weighted means","volume":"2016","author":"Sbert","year":"2016","journal-title":"J. Inequalities Appl."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"15","DOI":"10.1186\/s13634-018-0531-2","article-title":"Multiple importance sampling revisited: Breaking the bounds","volume":"2018","author":"Sbert","year":"2018","journal-title":"EURASIP J. Adv. Signal Process."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"461","DOI":"10.1007\/s11222-008-9089-4","article-title":"Adaptive methods for sequential importance sampling with application to state space models","volume":"18","author":"Cornebise","year":"2008","journal-title":"Stat. Comput."},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"M\u00edguez, J. (2017, January 23\u201325). On the performance of nonlinear importance samplers and population Monte Carlo schemes. Proceedings of the 2017 22nd International Conference on Digital Signal Processing (DSP), London, UK.","DOI":"10.1109\/ICDSP.2017.8096057"},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"10","DOI":"10.1109\/LSP.2013.2288355","article-title":"On the chi square and higher-order chi distances for approximating f-divergences","volume":"21","author":"Nielsen","year":"2014","journal-title":"IEEE Signal Process. Lett."},{"key":"ref_32","unstructured":"Geyer, C.J. (1991). Reweighting Monte Carlo Mixtures, University of Minnesota. Technical Report."}],"container-title":["Entropy"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1099-4300\/24\/2\/191\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T22:09:19Z","timestamp":1760134159000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1099-4300\/24\/2\/191"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,1,27]]},"references-count":32,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2022,2]]}},"alternative-id":["e24020191"],"URL":"https:\/\/doi.org\/10.3390\/e24020191","relation":{},"ISSN":["1099-4300"],"issn-type":[{"value":"1099-4300","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,1,27]]}}}