Summer Lu

adu187@r.postjobfree.com 240-***-**** Washington, D.C. http://www.linkedin.com/in/tonglu-math Education

Ph.D. Candidate in Mathematical Statistics University of Maryland, College Park May 2023 GPA: 3.92 / 4.00 Concentration: Statistics and Machine Learning

• Maryland Scholars Award Graduate Committee Statistics Student Representative Dean’s Fellowship Recipient Aziz Osborn Gold Medal in Teaching Excellence

B.S. with Highest Distinction in Mathematics University of Maryland, College Park May 2018

• Major GPA: 3.95 / 4.00 Concentration: Statistics

• Senior Marshal Appointed by President UMD Mathematics Competition 2016 Winner UMD Academic Excellence Award Work Experience

Data Scientist Intern, PhD Google May 2022- Aug. 2022

• Developed causal inference models and validation methodologies to assess impacts of multi-level interventions (e.g., feature launches, recommendation adoptions) on Google Cloud‘s overall revenue (sample size: 2 million)

• Built and leveraged ML models (Gradient Boosting, XGBoost, Ordered Logistic Regression) to estimate heterogeneous causal effects on top Cloud products’ revenue

• Programmed production code that generated real-time revenue impact analysis for Google Cloud products Machine Learning Researcher Center for Advanced Imaging Research, UMD May 2021- May 2022

• Developed, deployed, and maintained deep-learning pipelines (CNNs, RNNs) for advanced brain imaging processing with team of 5 that boosted accuracy by 32% and efficiency by 45% (data dimension: 20TB)

• Built classification and predictive models to evaluate and predict inter-subject brain entropy progression trajectory

• Developed various ML models to characterize transfer-entropy-based information flow between brain regions

• Led project work and provided internal statistical consulting and review for lab team Lecturer Mathematical Department & Academic Achievement Program, UMD Summer, 2017- 2021

• Developed instructional and diagnostic materials for Calculus I, II and Statistics I, II

• Presented lectures, evaluated assignments, and conducted office hours Research Experience

Research Assistant Mathematical Department & Maryland Psychiatric Research Center Jointly, UMD Sep. 2018- Present

• Develop ML and statistical network models to reveal abnormal spatiotemporal patterns in neuroimaging and genetics data via novel loss functions, spectral clustering, graph theory, time series analysis, and non-convex optimization

• Construct study design and develop statistical methodologies for longitudinal clinical trials seeking new drug treatments

• Establish network-level multivariate statistical inference methods to infer phenotypical spatiotemporal patterns

• Craft rich data visualizations and present to audiences with broad expertise at various conference meetings Applied Scientist Inmas Program, Johns Hopkins University Sep. 2021- Mar. 2022

• Admitted to highly selective Ph.D. data science bootcamps and collaborated intensively on challenging projects

• Designed, experimented, and evaluated advanced ML/statistical models and computational tools to tackle different industry's pressing problems (using LightGBM, Naïve Bayes, SVM, etc.)

• Brainstormed quantitative and strategic decisions and presented integrated progress regularly Technical Capabilities

• Coding: Python, R, SQL, MATLAB, SAS, TensorFlow, PyTorch, Hadoop, Java, C++, Scala, Spark, Ruby, Tableau, Git

• Methodologies: machine learning, statistics, A/B testing and experiment, causal inference, data structures & algorithms, artificial intelligence, scalable datastores, large consortium datasets management and analysis Publications

• Mo, Ye, Ke, Lu, T., Canida, T., Liu, S., ... & Chen, S. (2021). A new Mendelian Randomization method to estimate causal effects of multivariable brain imaging exposures. Pacific Symposium on Biocomputing 2022 (pp. 73-84). (+ Equal contributor)

• Lu, T., Zhang Y., Kochunov, P., Hong, E., Chen, S. (2022+) Statistical Network Model for Voxel-Pair Level Brain Connectivity Analysis with Spatial Contiguity Constraints. Under revision in Annals of Applied Statistics.

• Lu, T., Huang, H., Chen, S. (2023+) Novel Stochastic Imputation Method of Missing Imaging Data in Whole Brain Analysis. Manuscript in preparation.

• Lu, T., Chen, S. (2023+) Altered Brain Connectome Study With K-Partite Graph Topology and Spectral Clustering. Manuscript in preparation.

• Zhang, L., Xie, D., Camargo, A., Song, D., Lu, T., ... &Wang, Z. (2021). Improving Sensitivity of Arterial Spin Labeling Perfusion MRI in Alzheimer's Disease Using Transfer Learning of Deep Learning Based ASL Denoising. Journal of Magnetic Resonance Imaging.

• Lu, T., Gary, A., Goldman, W. (2017) Möbius Transformations Acting on Chains Generated by Hermitian Matrices in Complex Plane and Unit Spheres. Geometry Labs United 2017 Conference.

Contact this candidate