Dehan Zhu

R&D in New Machine Learning & Stochastic Optimization Methods for High-dimensional Modeling Improved Transdimensional Markov chain Monte Carlo (MCMC) method. Stochastic Boltzmann machine learning and Bayesian modeling. Other projects using machine learning methods: Random Forest, Extreme Gradient Boosting, Support Vector Machine, Decision Tree, Regularized Regression, Neural Network. Statistical analysis: hypothesis testing, hierarchical modeling, PCA, uncertainty/risk analysis.

3D Data Processing, Visualization & Data Analysis 1. Analyzed & processed 3D large data of seismic signals (2D spatial + 1D time series). 2. Performed image processing to achieve high-resolution visualization for the subsurface. 3. Invented a new method potentially integrated in future versions of the patented technology. 4. Created technical documentation for training users and clients.

Data Analysis & Visualization to Optimize Multi-million Dollar Drilling Projects in Oil/gas Business 1. Applied new technology to facilitate decision-making by analyzing 3D seismic big data. 2. Feedback to research team to improve the Shell inhouse technology.

R&D in Data Science and Machine Learning. Application in Geophysics.