Chinese University of Hong Kong
Grades
cGPA: 3.95/4.0
Relevant Modules
Speech and Language Processing: Introduced the underlying statistical approaches and major modelling techniques used in automatic speech recognition (ASR). The coursework was building GMM-HMM or DNN-HMM based speech recognition systems
Foundations of Optimization: Basic analysis on various optimization problems
University of Cambridge
Grades
First year: 2.i ranking: 47%;
Second year: 2.i, ranking: 38%;
Third year: Pass1 with Exhibition prize;
Fourth year: Project: 2.i, Examinations: 2.ii
Relevant Modules
Statistical Signal Processing: Markov chains; Auto-correlation functions; Wide sense stationary; Time series models, etc.
Inference, Probabilistic Machine Learning: Bayesian inference; Gaussian process; Expectation Maximization (EM) algorithm; Basic graphical models, etc.
Deep Learning and Structured Data: Introduced commonly used machine learning and deep learning models: Support Vector Machines, Ensemble models, Gaussian mixture models, HMM; CNN, RNN, LSTM, Transformer, etc.
Coursework:
Logistic Classification Model: Used logistic regression to do binary classification on a dataset and improved the model by radial basis functions
Retinal Ganglion Cell Model: Used sparse PCA to extract visual features from images, to imitate the way retinal ganglion cell works.
Probabilistic Machine Learning Coursework: Trained a Bayesian Gaussian Processes regression model for given data