Federated feature selection for cyber-physical systems of systems
Cassarà P., Gotta A., Valerio L.
Computer Science - Machine Learning Feature selection Human State Monitoring Artificial Intelligence Computer Science - Networking and Internet Architecture Computer Networks and Communications Internet of Things Machine Learning Feature Selection Electrical and Electronic Engineering Federated learning Distributed learning FOS: Computer and information sciences Aerospace Engineering Automotive Engineering Federated Learning Networking and Internet Architecture (cs.NI) Autonomous System Machine Learning (cs.LG) Mutual information
Autonomous vehicles (AVs) generate a massive amount of multi-modal data that once collected and processed through Machine Learning algorithms, enable AI-based services at the Edge. In fact, only a subset of the collected data present infor- mative attributes to be exploited at the Edge. Therefore, extracting such a subset is of utmost importance to limit computation and com- munication workloads. Doing that in a distributed manner imposes the AVs to cooperate in finding an agreement on which attributes should be sent to the Edge. In this work, we address such a problem by proposing a federated feature selection (FFS) algorithm where the AVs collaborate to filter out, iteratively, the less relevant at- tributes in a distributed manner, without any exchange of raw data, thought two different components: a Mutual-Information-based feature selection algorithm run by the AVs and a novel aggregation function based on the Bayes theorem executed on the Edge. The FFS algorithm has been tested on two reference datasets: MAV with images and inertial measurements of a monitored vehicle, WESAD with a collection of samples from biophysical sensors to monitor a relative passenger. The numerical results show that the AVs converge to a minimum achievable subset of features with both the datasets, i.e., 24 out of 2166 (99%) in MAV and 4 out of 8 (50%) in WESAD, respectively, preserving the informative content of data.
Source: IEEE transactions on vehicular technology 71 (2022): 9937–9950. doi:10.1109/TVT.2022.3178612
Publisher: Institute of Electrical and Electronics Engineers,, New York , Stati Uniti d'America
[2] L.-l. Chen, Y. Zhao, P.-f. Ye, J. Zhang, and J.-z. Zou, “Detecting driving stress in physiological signals based on multimodal feature analysis and kernel classifiers,” Expert Systems with Applications, vol. 85, pp. 279- 291, 2017.
[3] Y. Saeys and P. Inza, I.and Larranaga, “A review of feature selection techniques in bioinformatics,” Bioinformatics Review, vol. 23, no. 19, pp. 2507-2517, August 2007.
[4] S. Egea, A. R. Manez, B. Carro, A. Sanchez-Esguevillas, and J. Lloret, “Intelligent iot traffic classification using novel search strategy for fastbased-correlation feature selection in industrial environments,” IEEE Internet of Things Journal, vol. 5, no. 3, pp. 1616-1624, June 2018.
[5] G. Brown, “A new perspective for information theoretic feature selection,” in Int. Conf. on Artificial Intelligence and Statistics, Clearwater Beach, Fl, USA, April 2009.
[6] X. Nguyen, J. Chan, S. Romano, and J. Bailey, “Effective Global Approaches for Mutual Information Based Feature Selection,” in ACM Int. Conf. on Knowledge Discovery and Data Mining, New York City, NY, USA, August 2014, pp. 1-10.
[7] D. Rubinstein, R.Y. Kroese, The Cross-Entropy Method: a Unified Approach to Combinatorial Optimization, Monte-Carlo Simulation, Machine Learning. Springer, 2004.
[8] K. Jovic´, A.and Brkic´ and N. Bogunovic´, “A review of feature selection methods with applications,” in IEEE Int. Conf. MIPRO 2015. IEEE, 2015, pp. 1200-1205.
[9] B. Venkatesh and J. Anuradha, “A Review of Feature Selection and Its Methods,” CYBERNETICS AND INFORMATION TECHNOLOGIES, vol. 19, no. 1, pp. 3-26, February 2019.
[10] F. Chandrashekar, G. Sahin, “A Survey on Feature Selection Methods,” Computers and Electrical Engineering, vol. 40, no. 1, pp. 16 - 28, 2014.
[11] L. Miao, J. Niu, “A Survey on Feature Selection,” in Elsevier Int. Conf. on Information Technology , Quantitative Management, Asan, South Korea, October 2016, pp. 919-926.
[12] R. Petroccia, P. Cassara`, and K. Pelekanakis, “Optimizing adaptive communications in underwater acoustic networks,” in OCEANS 2019 MTS/IEEE SEATTLE. IEEE, 2019, pp. 1-7.
[13] G. F. Anastasi, P. Cassara`, P. Dazzi, A. Gotta, M. Mordacchini, and A. Passarella, “A hybrid cross-entropy cognitive-based algorithm for resource allocation in cloud environments,” in 2014 IEEE Eighth International Conference on Self-Adaptive and Self-Organizing Systems. IEEE, 2014, pp. 11-20.
[14] F. Guarino, P. Cassara`, S. Longo, M. Cellura, and E. Ferro, “Load match optimisation of a residential building case study: A cross-entropy based electricity storage sizing algorithm,” Applied energy, vol. 154, pp. 380- 391, 2015.
[15] J. Konecˇny´ , B. McMahan, and D. Ramage, “Federated Optimization:Distributed Optimization Beyond the Datacenter,” arXiv Prepr. arXiv1511.03575, no. 1, pp. 1-5, 2015.
[16] H. B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. y. Arcas, “Communication-Efficient Learning of Deep Networks from Decentralized Data,” in Proc. the20thInternational Conf. Artif. In-telligence Stat., vol. 54, 2017, pp. 248-50.
[17] S. Wang, T. Tuor, T. Salonidis, K. K. Leung, C. Makaya, T. He, and K. Chan, “When Edge Meets Learning: Adaptive Control for ResourceConstrained Distributed Machine Learning,” Proc. - IEEE INFOCOM, vol. 2018-April, pp. 63-71, 2018.
[18] M. M. Amiri and D. Gunduz, “Machine Learning at the Wireless Edge: Distributed Stochastic Gradient Descent Over-the-Air,” no. Ml, pp. 1-12, 2019.
[19] S. P. Karimireddy, S. Kale, M. Mohri, S. J. Reddi, S. U. Stich, and A. T. Suresh, “SCAFFOLD: Stochastic controlled averaging for ondevice federated learning,” arXiv, pp. 1-30, 2019.
[20] M. Mohri, G. Sivek, and A. T. Suresh, “Agnostic Federated Learning,” pp. 1-30, 2019.
[21] Y. Mao, “Learning from Differentially Private Neural Activations with Edge Computing,” 2018 IEEE/ACM Symp. Edge Comput., pp. 90-102, 2018.
[22] V. Mothukuri, R. M. Parizi, S. Pouriyeh, Y. Huang, A. Dehghantanha, and G. Srivastava, “A survey on security and privacy of federated learning,” Future Generation Computer Systems, vol. 115, pp. 619 - 640, 2021.
[23] L. Valerio, A. Passarella, and M. Conti, “Hypothesis transfer learning for efficient data computing in smart cities environments,” in International Conference on Smart computing (SMARTCOMP 2016), St. Louis, Missouri, IEEE, Ed., 2016.
[24] --, “A communication efficient distributed learning framework for smart environments,” Pervasive and Mobile Computing, vol. 41, pp. 46 - 68, 2017.
[25] X. Ye, H. Li, A. Imakura, and T. Sakurai, “Distributed Collaborative Feature Selection Based on Intermediate Representation,” in In Proc. of Int. Conf. IJCAI 2019, 2019, pp. 4242-4149.
[26] P. Schmidt, A. Reiss, R. Duerichen, C. Marberger, and K. Van Laerhoven, “Introducing wesad, a multimodal dataset for wearable stress and affect detection,” in ACM Int. Conf. ICMI 2018. ACM, 2019, pp. 1-9.
[27] G. Brown, A. Pocock, M. Zhao, and M. Zheng, “Conditional likelihood maximisation: A unifying framework for information theoretic feature selection,” Springer Journal of Machine Learning Research, vol. 13, pp. 27-66, 2012.
[28] R. McEliece, In The Theory of Information and Coding: A Mathematical Framework for Communication, ser. Encyclopedia of Mathematics and Its Applications. MA, USA: Addison-Wesley Publishing Company: Reading, 1977, vol. Vol. 3.
[29] J. Cover, T.M. Thomas, Elements of Information Theory. New York, NY, USA: John Wiley and Sons, Inc, 1991.
[30] A. Chaovalitwongse, I. Androulakis, and P. Pardalos, “Quadratic Integer Programming: Complexity and Equivalent Form,” Springer Encyclopedia of Optimization, pp. 3153-3159, September 2009.
[31] P. Chen, C. Huang, C. Lien, and Y. Tsai, “An efficient hardware implementation of hog feature extraction for human detection,” IEEE Transactions on Intelligent Transportation Systems, vol. 15, no. 2, pp. 656-662, 2014.
[32] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. The MIT Press, 2016.
[33] A. Costa, O. D. Jones, and D. Kroese, “Convergence properties of the cross-entropy method for discrete optimization,” Elsevier Operation Research Letters, vol. 35, no. 7, pp. 573-580, September 2007.
[34] K. Knopp, Infinite Sequences And Series. Dover Publications Inc., 1956.
Metrics
Back to previous page
@article{oai:it.cnr:prodotti:471809,
title = {Federated feature selection for cyber-physical systems of systems},
author = {Cassarà P. and Gotta A. and Valerio L.},
publisher = {Institute of Electrical and Electronics Engineers,, New York , Stati Uniti d'America},
doi = {10.1109/tvt.2022.3178612 and 10.5281/zenodo.6901226 and 10.48550/arxiv.2109.11323 and 10.5281/zenodo.6901227},
journal = {IEEE transactions on vehicular technology},
volume = {71},
pages = {9937–9950},
year = {2022}
}
0000-0002-3704-4133ISTI Repository
Postprint version
Published version
10.1109/tvt.2022.3178612
10.5281/zenodo.6901226
10.48550/arxiv.2109.11323
10.5281/zenodo.6901227
TEACHING
A computing toolkit for building efficient autonomous applications leveraging humanistic intelligence
HumanE-AI-Net
HumanE AI Network
MARVEL
Multimodal Extreme Scale Data Analytics for Smart Cities Environments
SoBigData-PlusPlus
SoBigData++: European Integrated Infrastructure for Social Mining and Big Data Analytics
