Nonlinear system identification using a recurrent network in a Bayesian framework
| العنوان: | Nonlinear system identification using a recurrent network in a Bayesian framework |
|---|---|
| المؤلفون: | Pietro Portolani, Matteo Matteucci, Stefano Spinelli, Alessandro Brusaferri |
| المصدر: | Industrial Informatics (INDIN), pp. 319–324, 23/07/2019-25/07/2019 info:cnr-pdr/source/autori:Brusaferri, A.; Matteucci, M.; Portolani, P.; Spinelli, S./congresso_nome:Industrial Informatics (INDIN)/congresso_luogo:/congresso_data:23%2F07%2F2019-25%2F07%2F2019/anno:2019/pagina_da:319/pagina_a:324/intervallo_pagine:319–324 INDIN |
| بيانات النشر: | IEEE, New York, USA, 2019. |
| سنة النشر: | 2019 |
| مصطلحات موضوعية: | 0209 industrial biotechnology, Computer science, Feature vector, Bayesian probability, Recurrent Neural Network, Inference, 02 engineering and technology, 010501 environmental sciences, Machine learning, computer.software_genre, 01 natural sciences, 020901 industrial engineering & automation, Control, Nonlinear systems, 0105 earth and related environmental sciences, Nonlinear system identification, business.industry, Probabilistic logic, System identification, System Identification, Nonlinear system, Recurrent Neural Net-work, Recurrent neural network, Artificial intelligence, business, computer, Simulation, System Identification, Recurrent Neural Network, Nonlinear systems, Simulation, Control |
| الوصف: | Modern deep neural networks are being widely exploited to solve challenging learning tasks, including nonlinear system identification. Bayesian system identification intrinsically encapsulate uncertainty in model parameters and provides forecasting distribution enabling enhanced analysis, simulation and control system design. Nevertheless, the application of the full Bayesian approach to articulated models as deep neural networks results quite challenging in practice. In this work we propose an identification technique for nonlinear dynamic systems exploiting a deep recurrent neural network with Long-Short Term Memory (LSTM) units retaining a Bayesian framework. To such an aim, we stacked the recurrent neural network with a probabilistic layer, decomposing the nonlinear dynamic model into a combination of flexible functions. Hence, deterministic and stochastic layers are trained jointly, forcing the learning algorithm to transform the input data sequences into a deterministic feature space encoded by the LSTM, useful for predictions. Besides, we deployed a scalable technique based on Variational Inference to deal with the exact inference intractability. We show the effectiveness of the proposed approach by the application to a widely exploited open benchmark for nonlinear system identification. |
| اللغة: | English |
| URL الوصول: | https://explore.openaire.eu/search/publication?articleId=doi_dedup___::4456c4bcf08227a2cd1169989c69511b https://publications.cnr.it/doc/406325 |
| حقوق: | RESTRICTED |
| رقم الأكسشن: | edsair.doi.dedup.....4456c4bcf08227a2cd1169989c69511b |
| قاعدة البيانات: | OpenAIRE |
| الوصف غير متاح. |