Yiming Liu Resume

Irvine, CA, *****

Æ 206-***-**** Q ac0xcf@r.postjobfree.com

//www.linkedin.com/in/yimingliu1992/

Education

+

University of California, Irvine Irvine, US

MSc in Statistics, Cum GPA: 3.87(current) Jun, 2015–Jun, 2017

+

University of Liverpool Liverpool, UK

BSc in Mathematics with Finance, First-Class Honours Degree Jun, 2015 Technical skills

+ Programming Languages and Softwares: Python (Numpy, Pandas, Matplotlib, Scikit-learn, Tensorflow), Matlab, R, SQL (MySQL, PostgreSQL), Hadoop Ecosystem (HDFS, Mapreduce, Pig, Hive), Latex, Tableau

+ Machine Learning: Regression, Classification, Clustering, Feature Engineering, Deep Learning( ANN, CNN),

+ Statistical Methods: Generalized Linear Model, Maximum Likelihood Estimation, Hypothesis Testing, Frequentist Inference, Numerical Optimization, Finite Mixture Models, Time Series Analysis, Bayesian Inference, MCMC Sampling, Variational Bayes, Bootstrap, Dimensionality Reduction(PCA, FA, t-SNE) Projects

’Deep Dream Image Generation’ May, 2017–Jun, 2017

+ Implemented Deep Dream Algorithm based on a pre-trained 22-layer GoogLeNet trained on ImageNet data set, using tensorflow.

+ Experimented various ways, such as taking a subset of feature maps and using guide images, to e ectively regularize the dreaming process in order to reduce cluttering of undesired features.

+ Generated new images by combining the content and style of two images through guided deep dream.

’Chaotic Hamiltonian Monte Carlo Sampling’ Feb, 2017–Mar, 2017

+ Implemented, using python, an adaptation of traditional HMC algorithm, which exploits the freedom in choosing momentum distribution to create chaotic sampling trajectories, on simulated 30-dimensional Gaussian with various covariance structures.

+ Compared the sampling results of CHMC to traditional HMC, revealing that CHMC produced samples with significantly less correlation, and its performance also exhibits less reliance on the choices of step-size during Leapfrog integration.

’Rainfall Prediction With Boosted Decision Trees’ Nov, 2016–Dec, 2016

+ Predicted the probability of rainfall at a location, based on (processed) infrared satellite image information.

+ Trained predictive Models based on boosting(Adaboost, Gradient boosting) of basic decision trees, using Scikit-learn and XGBoost.

+ Tuned model parameters through 3-fold cross validation and randomized search, and final model obtained a 0.8 AUC score on a separate test set.

’Classification Of Higgs Boson Tau-Tau Decays With Bagged Dropout Neural Networks’ Feb, 2016–Mar, 2016

+ Distinguished the signals of Higgs-Boson to Tau-Tau decays from severe background noise based on data from the Atlas experiment.

+ Performed dimensionality reduction through principal component analysis to extract most prominent signals.

+ Trained bootstrap aggregated 3-hidden-layer feed-forward artificial neural networks with dropout regularization on training set, using R’s H2O library, and the optimal model chosen, based on a 3-fold cross-validation, obtained about 85 percent classification correction rate on a separate test set.

Work Experience

Industrial And Commercial Bank Of China Wuxi

Statistician Intern Jun 2014–Sept 2014

+ Built up predictive models to estimate the default probabilities of certain clients with penalized regression, helping minimize, for the bank, the probability of potential financial losses due to insolvency or bankruptcy of clients.

+ Established forecast about the earnings of multiple financial portfolios weekly with GARCH based model, generating statistical evidence for proposed investment strategies.

+ Recorded and queried the credit scores of dozens of clients daily, facilitating proper information flowwithin a group of 12 people.

Contact this candidate