Abstract
The paper describes a novel methodology to compare learning algorithms by exploiting their performance maps. A performance map enhances the comparison of a learner across learning contexts and it also provides insights in the distribution of a learners’ performances across its parameter space. Also some initial empirical findings are commented.
In order to explain the novel comparison methodology, this study introduces the notions of learning context, performance map, and high performance function. These concepts are then applied to a variety of learning contexts to show how the methodology can be applied.
Finally, we will use meta-optimization as an instrument to improve the efficiency of the parameter space search with respect to its complete enumeration. But, note that meta-optimization is neither an essential part of our methodology nor the focus of our study.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
We projected two parameters on the X axis for creating the 3-d graphs. In particular, we projected ‘min impurity’ and ‘min samples’ on the X axis and ‘max depth’ on the Y axis for DT. Then the label ‘0.1 - 20’ on the X axis has to be interpreted as ‘min impurity’ = 0.2 and ‘min samples = 20’. Instead, for SVM, we projected ‘gamma’ and ‘C value’ on the X axis and ‘kernel’ on the Y axis.
References
Bergmeir, C., Benítez, J.M.: On the use of cross-validation for time series predictor evaluation. Inf. Sci. 191, 192–213 (2012)
Blum, C., Roli, A.: Metaheuristics in combinatorial optimization: overview and conceptual comparison. ACM Comput. Surv. 35(3), 268–308 (2003)
Bradley, A.P.: The use of the area under the roc curve in the evaluation of machine learning algorithms. Pattern Recogn. 30(7), 1145–1159 (1997)
Camilleri, M., Neri, F.: Parameter optimization in decision tree learning by using simple genetic algorithms. WSEAS Trans. Comput. 13, 582–591 (2014)
Camilleri, M., Neri, F., Papoutsidakis, M.: An algorithmic approach to parameter selection in machine learning using meta-optimization techniques. WSEAS Trans. Syst. 13(1), 203–212 (2014)
Cortes, C., Vapnik, V.: Support-vector networks. Mach. Learn. 20(3), 273–297 (1995)
Eiben, A., Hinterding, R., Michalewicz, Z.: Parameter control in evolutionary algorithms. IEEE Trans. Evol. Comput. 3(2), 124–141 (1999)
Feurer, M., Hutter, F.: Hyperparameter optimization. In: Hutter, F., Kotthoff, L., Vanschoren, J. (eds.) Automated Machine Learning. TSSCML, pp. 3–33. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-05318-5_1
García-Margariño, I., Plaza, I., Neri, F.: ABS-mindburnout: an agent-based simulator of the effects of mindfulness-based interventions on job burnout. J. Comput. Sci. 36, 101012 (2019)
Goldberg, D.: Genetic Algorithms in Search, Optimization, and Machine Learning. Addison-Wesley, Reading, Ma (1989)
Grefenstette, J.J.: Optimization of control parameters for genetic algorithms. IEEE Trans. Syst. Man Cybern. 16(1), 122–128 (1986)
Lorenzo, P.R., Nalepa, J., Kawulok, M., Ramos, L.S., Pastor, J.R.: Particle swarm optimization for hyper-parameter selection in deep neural networks. In: Proceedings of the Genetic and Evolutionary Computation Conference, pp. 481–488. GECCO 2017. ACM (2017)
Marino, A., Neri, F.: PID tuning with neural networks. In: Nguyen, N.T., Gaol, F.L., Hong, T.-P., Trawiński, B. (eds.) ACIIDS 2019. LNCS (LNAI), vol. 11431, pp. 476–487. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-14799-0_41
Neri, F.: Traffic packet based intrusion detection: decision trees and genetic based learning evaluation. WSEAS Trans. Comput. 4(9), 1017–1024 (2005)
Neri, F.: PIRR: a methodology for distributed network management in mobile networks. WSEAS Trans. Inf. Sci. Appl. 5(3), 306–311 (2008)
Neri, F.: Learning and predicting financial time series by combining natural computation and agent simulation. In: Di Chio, C., et al. (eds.) EvoApplications 2011. LNCS, vol. 6625, pp. 111–119. Springer, Heidelberg (2011). https://doi.org/10.1007/978-3-642-20520-0_12
Neri, F.: Agent-based modeling under partial and full knowledge learning settings to simulate financial markets. AI Commun. 25(4), 295–304 (2012)
Neri, F.: A comparative study of a financial agent based simulator across learning scenarios. In: Cao, L., Bazzan, A.L.C., Symeonidis, A.L., Gorodetsky, V.I., Weiss, G., Yu, P.S. (eds.) ADMI 2011. LNCS (LNAI), vol. 7103, pp. 86–97. Springer, Heidelberg (2012). https://doi.org/10.1007/978-3-642-27609-5_7
Neri, F.: Learning predictive models for financial time series by using agent based simulations. In: Nguyen, N.T. (ed.) Transactions on Computational Collective Intelligence VI. LNCS, vol. 7190, pp. 202–221. Springer, Heidelberg (2012). https://doi.org/10.1007/978-3-642-29356-6_10
Neri, F.: Coevolution and learning symbolic concepts: statistical validation: empirical statistical validation of co-evolutive machine learning systems. In: Neri, F. (ed.) 7th International Conference on Machine Learning Technologies ICMLT 2022, pp. 244–248. ACM (2022)
Neri, F.: Explainability and interpretability in agent based modelling to approximate market indexes. In: Neri, F. (ed.) 8th International Conference on Machine Learning Technologies ICMLT 2023, in press. ACM (2023)
Neri, F.: Explainability and interpretability in decision trees and agent based modelling when approximating financial time series, a matter of balance with performance. In: Neri, F. (ed.) 2023 8th International Conference on Computational Intelligence and Applications ICCIA 2023, in press. Springer press (2023)
Neri, F., Margariño, I.: Simulating and modeling the DAX index and the USO Etf financial time series by using a simple agent-based learning architecture. Expert Syst. 37(4), 12516 (2020)
Neri, F.: Software agents as a versatile simulation tool to model complex systems. WSEAS Trans. Info. Sci. and App. 7(5), 609–618 (2010)
Neri, F.: Case study on modeling the silver and nasdaq financial time series with simulated annealing. In: Rocha, Á., Adeli, H., Reis, L.P., Costanzo, S. (eds.) WorldCIST’18 2018. AISC, vol. 746, pp. 755–763. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-77712-2_71
Neri, F.: Combining machine learning and agent based modeling for gold price prediction. In: Cagnoni, S., Mordonini, M., Pecori, R., Roli, A., Villani, M. (eds.) WIVACE 2018. CCIS, vol. 900, pp. 91–100. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-21733-4_7
Neri, F.: How to identify investor’s types in real financial markets by means of agent based simulation. In: 6th International Conference on Machine Learning Technologies ICMLT 2021, pp. 144–149. ACM (2021)
Neri, F.: Unpublished result: Domain specific concept drift detectors for predicting financial time series. https://arxiv.org/abs/2103.14079. Arxiv.org (UP)
Pedregosa, F., et al.: Scikit-learn: machine learning in Python. J. Mach. Learn. Res. 12, 2825–2830 (2011)
Quinlan, J.R.: C4.5: programs for machine learning. Morgan Kaufmann, California (1993)
Racine, J.: Consistent cross-validatory model-selection for dependent data: hv-block cross-validation. J. Econometr. 99(1), 39–61 (2000)
Refaeilzadeh, P., Tang, L., Liu, H.: Cross-Validation, pp. 532–538. Springer, US, Boston, MA (2009). https://doi.org/10.1007/978-0-387-39940-9_565
Reif, M., Shafait, F., Dengel, A.: Meta-learning for evolutionary parameter optimization of classifiers. Mach. Learn. 87(3), 357–380 (2012)
Schlimmer, J.C.: Concept acquisition through representational adjustment. Doctoral dissertation, Department of Information and Computer Science, University of California, Irvine, CA (1987)
Stone, M.: Cross-validatory choice and assessment of statistical predictions. discussion. J. Royal Statist. Soc. Ser. B 36, 111–147 (1974)
Wolpert, D., Macready, W.: No free lunch theorems for optimization. IEEE Trans. Evol. Comput. 1(1), 67–82 (1997)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Neri, F. (2023). Mapping Learning Algorithms on Data, a Promising Novel Methodology to Compare Learning Algorithms. In: Neri, F., Du, KL., Varadarajan, V., San-Blas, AA., Jiang, Z. (eds) Computer and Communication Engineering. CCCE 2023. Communications in Computer and Information Science, vol 1823. Springer, Cham. https://doi.org/10.1007/978-3-031-35299-7_18
Download citation
DOI: https://doi.org/10.1007/978-3-031-35299-7_18
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-35298-0
Online ISBN: 978-3-031-35299-7
eBook Packages: Computer ScienceComputer Science (R0)