Luong Ha Nguyen's Research Page

Former position

Postdoc | 2019 - 2021

Contact

luong-ha.nguyen@polymtl.ca

Education

Ph.D., Civil Engineering | 2017-2019
   Polytechnique Montreal
   Advisor: James-A Goulet       

M.Sc., Civil Engineering | 2015-2017
   Polytechnique Montreal
   Advisor: James-A Goulet       

Engineering Diploma, Civil Engineering | 2013-2016
   École Spéciale des Travaux Publics, du Bâtiment et de l’Industrie (ESTP)
   Paris, France       

B.ing., Civil Engineering | 2011-2013
   Rennes 1 University
   Rennes, France

Research

Project

Title: Bayesian Neural Networks for Time Series Analysis

Summary: My research focuses on developing the theory of Bayesian linear neural networks (BLNNs) for time series analysis. The BLNNs will allow modeling nonlinear relationships as well as complex patterns that commonly exhibit in time series. The expected outcome of this research is able to provide an open-source platform including generic and reliable methods for interpreting any time series in real time. This outcome will allow accelerating scientific development and transferring theoretical contributions to various practical applications in different domains.

Publications

[2021]  Analytically Tractable Hidden-States Inference in Bayesian Neural Networks

            Nguyen, L.H. and Goulet, J.-A.,

            arXiv:2107.03759

            [PDF] [ ] [ ] [ ]

[2021]  Analytically Tractable Bayesian Deep Q-Learning

            Nguyen, L.H. and Goulet, J.-A.,

            arXiv:2106.11086

            [PDF] [ ] [ ] [ ]

[2021]  Analytically Tractable Inference in Deep Neural Networks

            Nguyen, L.H. and Goulet, J.-A.,

            arXiv:2103.05461

            [PDF] [ ] [ ] [ ]

[2021]  Anomaly detection using state-space models and reinforcement learning

            Khazaeli, S., Nguyen, L.H. and Goulet, J.-A.

            Structural Control and Health Monitoring. Volume 28, Issue 6, e2720.

            [PDF] [EndNote] [BibTeX] [DOI link] [ ]

[2020]  Tractable Approximate Gaussian Inference for Bayesian Neural Networks

            Goulet, J.-A., Nguyen, L.H. and Amiri, S.

            arXiv:2004.09281

            [PDF] [ ] [ ] []

[2019]  Real-time Anomaly Detection in the Behaviour of Structures

            Nguyen, L.H.

            Ph.D. Thesis, Polytechnique Montréal

            [PDF] [EndNote] [BibTeX]

[2019]  Real-time Anomaly Detection with Bayesian Dynamic Linear Models

            Nguyen, L.H. and Goulet, J.-A.

            Structural Control and Health Monitoring. In press

            [PDF] [EndNote] [BibTeX] [DOI link]

[2019]  A Kernel-based Method for Modeling Non-Harmonic Periodic Phenomena in

            Bayesian Dynamic Linear Models

            Nguyen, L.H., Gaudot, I., Shervin Khazaeli and Goulet, J.-A.

            Frontiers in Built Environment. Vol.5, i.d.:434090

            [PDF] [EndNote] [BibTeX] [DOI link]

[2019]  Uncertainty quantification for model parameters and hidden state variables in

            bayesian dynamic linear models

            Nguyen, L.H., Gaudot, I., and Goulet, J.-A.

            Structural Control and Health Monitoring. Vol. 26, Issue 3, pp.e2136

            [PDF] [EndNote] [BibTeX] [DOI link]

[2018]  Anomaly Detection with the Switching Kalman Filter for Structural Health Monitoring

            Nguyen, L.H. and Goulet, J.-A.

            Structural Control and Health Monitoring. Vol. 24, Issue 4, pp.e2136

            [PDF] [EndNote] [BibTeX] [DOI link]

[2018]  Structural health monitoring with dependence on non-harmonic periodic hidden covariates
            Nguyen, L.H. and Goulet, J.-A.

            Engineering Structures, 166:187 – 194.

            [PDF] [EndNote] [BibTeX] [DOI link]