Peer-reviewed research by Trieu Hai Vo in deep learning for environmental time series, water-quality monitoring, and physics-informed machine learning.
2026
KDD
Depth-wise Multivariate Imputation for Environmental Time Series
Trieu H. Vo, Cuong V. Nguyen, Ana Sophia Cavalcanti, and Leonardo Bobadilla
Time series data from environmental monitoring systems often contain missing values caused by sensor failures, transmission errors, or maintenance interruptions. While existing imputation methods perform well in many cases, they often fail to account for the unique vertical correlations between adjacent depth layers in stratified environments such as lakes or oceans. To address this limitation, we propose Depth-wise Multivariate Imputation, a time series imputation method that can simultaneously model dependencies across depth, feature, and temporal dimensions, ensuring the imputation process explicitly accounts for the complex physical interactions within stratified environmental systems. We demonstrate the effectiveness of our method through experiments on several datasets, including a newly collected multi-depth water quality dataset from Miami, USA.
@unpublished{vo2026dmi,title={Depth-wise Multivariate Imputation for Environmental Time Series},author={Vo, Trieu H. and Nguyen, Cuong V. and Cavalcanti, Ana Sophia and Bobadilla, Leonardo},note={Manuscript under peer review at ACM SIGKDD 2026},year={2026},}
2025
ICLR-W
Lake Water Temperature Modeling Using Physics-Informed Neural Networks
Trieu H. Vo, Cuong V. Nguyen, Dongsheng Luo, and Leonardo Bobadilla
In ICLR 2025 Workshop on Tackling Climate Change with Machine Learning, 2025
Assessing water quality in bodies of water is important in evaluating the effects of climate change and its anthropogenic impacts. Such assessments often require good models of key indices such as water temperature, pH, or oxygen levels. In this work, we investigate time series models for lake water temperatures at multiple depths and develop a physics-informed neural network based on Koopman embeddings and LSTM that is capable of forecasting water temperatures in the long term. Experiment results show that our model can achieve a good performance and significantly outperforms the conventional LSTM model for this time series forecasting problem.
@inproceedings{vo2025lake,title={Lake Water Temperature Modeling Using Physics-Informed Neural Networks},author={Vo, Trieu H. and Nguyen, Cuong V. and Luo, Dongsheng and Bobadilla, Leonardo},booktitle={ICLR 2025 Workshop on Tackling Climate Change with Machine Learning},year={2025},url={https://www.climatechange.ai/papers/iclr2025/48},}
