Artificial intelligence is changing the landscape of many industries in amazing ways: transportation, healthcare, education, ... ... I am very interested in the fundamental theory of machine learning and its applications in earth science. If you are looking to collaborating with a researcher with experience in both physics-based numerical modeling and data-driven knowledge discovery, please feel free to contact me!
Below are some projects I have been working on, and you can find more about my recent work here.
Below are some projects I have been working on, and you can find more about my recent work here.
|
Machine Learning for Spatial Interpolation of Geo-science Data
The objective of this study is to develop a Clustering-Based Ensemble Interpolator (CBEI) to reconstruct river bathymetry from scattered data samples.In this study, random forest (RF) and support vector machine (SVM) are used to improve the conventional spatial interpolation methods. |
|
"Learn to Write Poems like Shakespeare"
In this project, I trained two unsupervised learning models to write poems using Shakespeare's poems. The first one is a Hidden Markov Model (HMM) trained on syllable-level, and the second one is a Recurrent Neural Network (RNN) trained on character-level. It has one layer of LSTM units, one dense layer and one output layer. |
|
Reinforcement Learning for Smart Cab
In this project, I implemented a Q-learning algorithm to navigate the smart cab through the simulated environment. The trained smart cab is able to make right decision based on the traffic lights and the intention of other cars at the crossing and arrive the destination with a high safety and reliability level. |