Export
Van-Hau Nguyen
· 1 previous affiliation
- Lecturer · Since 2003
Research Overview
Van-Hau Nguyen focuses on constraint programming, Boolean satisfiability (SAT) encodings, and applied machine learning for environmental forecasting. Early work explored the performance trade-offs between SAT and constraint programming approaches for combinatorial problems, specifically developing novel encodings for cardinality constraints. More recently, this trajectory has expanded to include collaborative applications of time-series forecasting and neural networks for coastal monitoring.
Research Trajectory
Van-Hau Nguyen’s research trajectory begins in the domain of combinatorial optimization, specifically focusing on the intersection and comparative performance of Boolean satisfiability (SAT) and constraint programming (CP). In 2014, Nguyen led an investigation into the practical trade-offs between these two paradigms. While SAT solvers traditionally operate as black-box systems and CP offers a more tunable approach, Nguyen's work in Solving the all-interval series problem: SAT vs CP provided a direct comparison using the all-interval series problem. The research demonstrated that applying a state-of-the-art SAT solver with an optimized encoding strategy could yield significantly higher performance than traditional CP methods for this specific problem class.
Building on this foundational work in SAT encodings, Nguyen continued to explore methods for translating practical constraints into propositional satisfiability instances. A major focus of this subsequent research was the at-most-one (AMO) constraint, a fundamental component in translating complex problems into SAT instances. As the first author of A New Method to Encode the At-Most-One Constraint into SAT (2015), Nguyen proposed the "AMO bimander encoding." This method was designed to be easily extensible to broader cardinality constraints commonly utilized in constraint programming. Crucially, the research proved that this new encoding allows unit propagation to achieve arc consistency, bridging an important technique from constraint programming into the SAT domain while maintaining competitive performance against other established encodings.
Following this period of focused work on algorithmic constraints and satisfiability, the available evidence indicates a shift toward applied machine learning and environmental monitoring. In 2021, Nguyen participated as a contributing co-author in a large collaborative study, Advanced Machine Learning Techniques for Predicting Nha Trang Shorelines. This research addressed the rapid coastal erosion in South Central Vietnam by applying statistical and machine learning models—including Seasonal Auto-regressive Integrated Moving Average (SARIMA), Neural Network Auto-Regression (NNAR), and Long Short-Term Memory (LSTM) networks—to forecast shoreline variations using surveillance camera imagery. In this context, Nguyen's involvement reflects an application of computational modeling techniques to complex, real-world environmental datasets.
Current Directions
Based on the most recent paper in the evidence from 2021, Nguyen's current research directions involve the application of advanced machine learning and time-series forecasting techniques to environmental management challenges. This work investigates how deep learning architectures, such as Long Short-Term Memory (LSTM) networks, and statistical models can be leveraged to process visual data from surveillance systems for predictive modeling. Rather than focusing purely on foundational algorithmic constraints as seen in earlier work, these recent efforts emphasize interdisciplinary collaboration to monitor and predict complex physical phenomena, such as coastal erosion and shoreline dynamics.
Profile updated
| Title | Venue | Year |
|---|---|---|
| Advanced Machine Learning Techniques for Predicting Nha Trang Shorelines | IEEE Access | 2021 |
| A New Method to Encode the At-Most-One Constraint into SAT | Soict | 2015 |
| Solving the all-interval series problem: SAT vs CP | Soict | 2014 |
Loading new papers…
Loading research neighbors…
See something wrong? Report a mistake