Abstract
In this paper we investigate the use of the Extreme Learning Machine (ELM) paradigm for the approximate minimization of a general class of functionals which arise routinely in operations research, optimal control and statistics problems. The ELM and, in general, neural networks with random hidden weights, have proved to be very efficient tools for the optimization of costs typical of machine learning problems, due to the possibility of computing the optimal outer weights in closed form. Yet, this feature is possible only when the cost is a sum of squared terms, as in regression, while more general cost functionals must be addressed with other methods. Here we focus on the gradient boosting technique combined with the ELM to address important instances of optimization problems such as optimal control of a complex system, multistage optimization and maximum likelihood estimation. Through the application of a simple gradient boosting descent algorithm, we show how it is possible to take advantage of the accuracy and efficiency of the ELM for the approximate solution of this wide family of optimization problems.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
We assume that the minimum exists. Otherwise, the problem can be redefined in terms of ε-optimal solutions.
References
Huang, G., Huang, G.-B., Song, S., You, K.: Trends in extreme learning machines: a review. Neural Netw. 61, 32–48 (2015)
Huang, G.-B., Cambria, E., Toh, K.-A., Widrow, B., Xu, Z.: New trends of learning in computational intelligence. IEEE Comput. Intell. Mag. 10, 16–17 (2015)
Huang, G.-B., Chen, L., Siew, C.-K.: Universal approximation using incremental constructive feedforward networks with random hidden nodes. IEEE Trans. Neural Netw. 17(4), 879–892 (2006)
Cervellera, C., Macciò, D.: Low-discrepancy points for deterministic assignment of hidden weights in extreme learning machines. IEEE Trans. Neural Netw. Learn. Syst. 27(4), 891–896 (2016)
Hastie, T., Tibshirani, R., Friedman, J.: The Elements of Statistical Learning: Data Mining, Inference and Prediction. Springer, New York (2009)
Mason, L., Baxter, J., Bartlett, P., Frean, M.: Boosting algorithms as gradient descent. In: Solla, S.A., Leen, T.K., Müller, K. (eds.) Advances in Neural Information Processing Systems, vol. 12, pp. 512–518. MIT Press, Cambridge (1999)
Freund, Y., Schapire, R.E.: A decision-theoretic generalization of on-line learning and an application to boosting. J. Comput. Syst. Sci. 55(1), 119–139 (1997)
Guo, H., Wang, J., Ao, W., He, Y.: SGB-ELM: an advanced stochastic gradient boosting-based ensemble scheme for extreme learning machine. Comput. Intell. Neurosci. 2018, 1–14, (2018)
Gelfand, I.M., Fomin, S.V.: Calculus of Variations. Prentice Hall, New Jersey (1963)
Zoppoli, R., Sanguineti, M., Parisini, T.: Approximating networks and extended Ritz method for the solution of functional optimization problems. J. Optim. Theory Appl. 112, 403–439 (2002)
Cervellera, C., Macciò, D., Muselli, M.: Functional optimization through semi-local approximate minimization. Oper. Res. 58(5), 1491–1504 (2010)
Alessandri, A., Cervellera, C., Macciò, D., Sanguineti, M.: Optimization based on quasi-Monte Carlo sampling to design state estimators for non-linear systems. Optimization 69(7), 963–984 (2010)
Gnecco, G., Sanguineti, M.: Neural approximations in discounted infinite-horizon stochastic optimal control problems. Eng. Appl. Artif. Intell. 74, 294–302 (2018)
Macciò, D., Cervellera, C.: Local kernel learning for data-driven control of complex systems. Expert Syst. Appl. 39(18), 13399–13408 (2012)
Fan, H., Tarun, P.K., Chen, V.: Adaptive value function approximation for continuous-state stochastic dynamic programming. Comput. Oper. Res. 40(4), 1076–1084 (2013)
Cervellera, C., Macciò, D.: F-discrepancy for efficient sampling in approximate dynamic programming. IEEE Trans. Cybern. 46, 1628–1639 (2016)
Bertsekas, D.: Dynamic Programming and Optimal Control, vol. I, 2nd edn. Athena Scientific, Belmont (2000)
Cervellera, C., Macciò, D.: A novel approach for sampling in approximate dynamic programming based on F-discrepancy. IEEE Trans. Cybern. 47(10), 3355–3366 (2017)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this chapter
Cite this chapter
Cervellera, C., Macciò, D. (2019). Gradient Boosting with Extreme Learning Machines for the Optimization of Nonlinear Functionals. In: Paolucci, M., Sciomachen, A., Uberti, P. (eds) Advances in Optimization and Decision Science for Society, Services and Enterprises. AIRO Springer Series, vol 3. Springer, Cham. https://doi.org/10.1007/978-3-030-34960-8_7
Download citation
DOI: https://doi.org/10.1007/978-3-030-34960-8_7
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-34959-2
Online ISBN: 978-3-030-34960-8
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)