James Cheng Peng Email & Phone Number

San Diego, California, United States

Affiliated to Dr. Axel Nimmerjahn Lab in Waitt Advanced Biophotonics Center

Arlington, Virginia, United States

Affiliated to Computational Bioinformatics and Bioimaging Lab

Baltimore, Maryland, United States

Teaching assistant for EN.553.663 (Network Models in Operations Research)

Baltimore, Md, United States

Project 1: Approximation to optimal strategy in Mozart-Café problemDesigned novel k-Markovian parameterization technique as generic representation for any symmetric rendezvous search strategy in Mozart-Café problemDeveloped Monte Carlo simulation programs by Python to estimate expected rendezvous time for any symmetric rendezvous search strategy parameterized by our k-Markovian techniqueUtilized Simultaneous Perturbation Stochastic Approximation (SPSA) algorithm on parameter optimization in k-Markovian modelling and reproduced optimal strategies in small size casesProject 2: Classical and model-free controllers in nonlinear problemsImplemented and evaluated powerful model-free controller and PID controller in driftless affine nonlinear discrete-time systems by PythonTransformed analysis of uncontrollability by PID into convergence analysis of nonlinear difference equationsProved sufficient conditions for such systems to be uncontrollable by PID, but controllable by model-free controllerProject 3: Reinforcement learning in sports schedulingCo-designed Markov Decision Process modelling for single round-robin tournament scheduling problemsDesigned innovative backtracking actions in action space, significantly increasing robustnessDeveloped custom reinforcement learning environment for round-robin tournament scheduling problems by OpenAI GymTrained reinforcement learning agents by ACKTR algorithm and successfully generated sample schedules with least number of breaks in single round-robin cases

Hong Kong

Co-designed and developed novel supervised learning model named Comonotone-Independence Bayesian Classifier (CIBer), whose core techniques include: 1. Utilizing hierarchical clustering algorithm to search for optimal partition of all predictive features 2. Using comonotonicity to model dependence structure and to estimate conditional joint probabilityCo-designed and implemented Joint Encoding algorithm used to encode multiple categorical variables by numerical values while preserving their internal relationshipCompared CIBer with classical supervised learning models including Logistic regression, SVM, Decision tree, MLP, KNN, and Naïve Bayes on several insurance-related datasets, obtaining out-of-sample accuracy over 10% higher than any other model and AUC values over 0.08 higher than any other model on every datasetExtended model with singleton classifier to more robust version with ensemble learning

Baltimore, Maryland, United States

Retrieved and processed terabytes of mobility data by Presto SQL from Cuebiq Workbench to help analyze impact of Ad Council’s COVID-19 vaccination education campaign on U.S. county-level weekly vaccination rateAnalyzed and visualized temporal relationship between over 15 million mobile App users’ exposure to vaccination advertisements and weekly vaccination rates at county-level by PythonReviewed literature about Bayesian hierarchical spatio-temporal model and processed data for model fittingConcluded that campaign reaching everyone would boost vaccination rate by 2.2% (95% uncertainty interval: 2.0% – 2.4%) on weekly basis, compared to baseline case of no campaign

Li Ka Shing Faculty Of Medicine

Managed all data modelling tasks in project about early detection of adolescent idiopathic scoliosis by infrared camera imagesUtilized synthetic minority oversampling technique to deal with unbalanced datasets of infrared camera images and increased F1 score in prediction on each level of scoliosis by over 0.1Trained and tested ensemble learning models on rebalanced dataset, demonstrating promising potential for use of infrared thermography to predict severity of patients’ scoliotic curves

Doctor Of Philosophy - Phd, Electrical And Computer Engineering

Affiliated to the Computational Bioinformatics and Bioimaging Lab. My research work bridges the fields of control & dynamical systems.

Master Of Science - Ms, Operations Research And Optimization

Bachelor'S Degree, Mathematics And Statistics

Exchange Study, Data Science

High School Diploma, Science, 679/750 In Gaokao 2016 At Hubei Province (Top 250 Among >180K Participants)