Potential Project Topics for the MS Paper

Currently Open Projects

In addition to topics closely related to their research areas, faculty members supervise Master's students on a wide range of topics. The following is a list of potential projects for MS papers proposed by each faculty.

Last update: 05/03/2024

  • Yali Amit: Image Analysis Using Statistical Models; Models and Simulations of Recurrent Networks of Neurons; Machine Learning and Prediction—Random Forests, Deep Convolutional Networks—Image Data, Speech Data, Marketing and Internet Data. Comparison of Biological Hebbian Learning and Machine Learning.
  • Mihai Anitescu: Large Scale Space-time Statistical Analysis of Physical Sciences Data; Optimization Under Uncertainty, with Applications Primarily to Energy Systems: Electrical Power Grid and Gas Networks; State-space Models of Time Series and Sequential Estimation.
  • Rina Foygel Barber:  Estimation, Inference, and Model Selection for High-dimensional Statistics; Convex and Nonconvex Optimization for High-dimensional Statistics; Multiple Testing; False Discovery Rate Control; Data Privacy; Low-rank Models.
  • Kendra Burbank:  Biologically plausible algorithms for visual object recognition; relationship between synaptic plasticity and Hebbian learning in deep networks.
  • Claire Donnat:  Bayesian Models for multimodal biomedical data analysis. Summarizing heterogeneous information: Meta-analysis. Knowledge graphs. Statistics on graphs and networks.
  • Chao Gao: Applied Network Analysis; Bayes Modeling and Variational Algorithms; Sports Data Analysis.
  • Gregory Lawler:  As long as it has something to do with Bridge, the card game that "the average defender operates in a fog of uncertainty."
  • Xinran Li:
  • Lek-Heng Lim:  Matrix theory/computations/applications: Topics from Stat 30900 or Stat 32940; Convex optimization/analysis: Topics from Stat 31015; Statistical tests of randomness on cellphone encryption schemes; Generalization of Shepp's self-normalized product.
  • Cong Ma:  Transfer learning (including covariate shift); applied and theoretical reinforcement learning; convex and nonconvex optimization for e.g., low-rank models; experiments related to self-supervised learning.
  • Peter McCullagh:  Statistics applied to epidemiology, health sciences, agricultural or environmental research, and so on; applications involving spatial structure, temporal structure, tree structure, and so on; asymptotic approximation of distributions; sparsity; random discrete structures such as random graphs, hypergraphs, trees and so on; dynamical random systems and neural networks.
  • Dan Nicolae:  Analysis of Microbiome Data; Latent Factor Models for Structured Response; Admixture Mapping for Expression Phenotypes; Modeling and Inference of Dynamical Networks; Prior Information in Subjective Bayesian Analysis.
  • John Reinitz:  Global Optimization by Stochastic Methods; Statistical Characterization of Fluctuations in Gene Expression; Stochastic Models of Gene Expression; Modeling and Inference of Dynamical Networks.
  • Aaron Schein:  Scalable Bayesian factor modeling of complex data, such as non-Gaussian data, time-series, networks, spatial data, multimodal data, high-dimensional discrete data; tensor decomposition; state-space models; causal inference from randomized experiments or panel data; applications in the social sciences.
  • Mary Silber:  Exploring Early Warning Signs of Tipping Points in Chaotic Time Series; Modeling Precipitation Time Series in Different Semi-Arid Regions of the Globe; Automated Detection and Quantification of Vegetation Patchiness from Satellite Image Data of Dryland Ecosystems.
  • Stephen Stigler:  Applied statistics in various fields.  Recent MS thesis titles: Predicting the Winners of NBA Games; Modeling Economic and Environmental Time Series in Three East Asian Countries; Pairs Trading Research—A Cointegration Approach; A Brief Investigation into Issues of Correlation in Fantasy Football; A Comparison of Techniques for Handling Missing Data; Combining Latent Topics with Document Attributes in Text Analysis.
  • Yi Sun:  Random matrix theory (free probability, exactly solvable models); neural networks (theory, data augmentation, robustness)
  • Victor Veitch:  
  • Jingshu Wang:  Deep learning methods for single-cell RNA-sequencing data; distribution deconvolution for understanding gene-gene co-bursting patterns; confounding adjustment of causal inference for genetic data; post-selection multiple hypotheses testing.
  • Mei Wang:  Statistical models and computational methods applied in scientific fields. Sample MS thesis topics in the past: Non-Gaussian Component Analysis, Dynamics and Community Detection of Knowledge Networks, Text Mining and Authorship Classification; Curve Clustering of Time Course Gene Expression Data Using Gaussian Processes; Statistical Credibility Theory and Its Applications to Actuarial Modeling; Permutation Tests for Small Samples.
  • Kirk Wolter:  Survey Design and Analysis for the Medicare Current Beneficiary Survey; Survey Design and Analysis for the National Immunization Survey; Survey Design and Analysis for the Survey of Doctoral Recipients.
  • Wei Biao Wu:  Value at Risk Backtest Methods; Regression Models with Correlated Errors; Estimating Market Index Using Time-varying Copulas; Large Scale Testing for Clustered Signals; GARCH Analysis with Generalized Methods of Moments.