New Undergraduate Course Offerings
Coming Soon! SDS 1610 AI for Everyone: Statistical Intuition and Societal Impacts
with Professor Joe Feldman
This course develops a foundational understanding of Artificial Intelligence and its societal impacts through statistical intuition and critical reasoning, without requiring rigorous mathematics or coding. The guiding questions throughout the course are: What data AI systems learn from, what their predictions mean, how systems are evaluated, and when do they fail?
SDS 2010 Analytical Tools for Statistics and Data Science I
with Professor José Figueroa-López and Professor Lizda Moncada
This is an introductory course in matrix algebra. The focus is on the concepts and computational aspects of matrix algebra using a high-level programming language or software, and its applications in Statistics and Data Science. Topics include the standard matrix and vector operations, inverses, determinants, system of linear equations, diagonalization (eigenvalues and eigenvectors), and other key matrix decompositions (LU, QR, and Cholesky). This is a short course (approx. 8 weeks).
SDS 2011 Analytical Tools for Statistics and Data Science II
with Professor José Figueroa-López and Professor Lizda Moncada
The course covers some topics of multivariate calculus that are particularly relevant for Statistics and Data Science. The focus is on the ideas, concepts, and computational implementation of the theory. Whenever possible, we will show the use of mathematical and statistical software (e.g., Matlab, Maple, and R) to obtain symbolic, exact, or approximate solutions of the mathematical tools developed in class. Topics include generalities on functions in several variables (graphs, level curves, limits, continuity), linear and quadratic functions in two variables, partial differentiation, multivariate Chain Rule, multivariate Taylor’s approximation, local and global extrema, Lagrange multipliers, multiple integration, change of variables formula (Jacobians), and polar, cylindrical, and spherical coordinates. This is a short course (approx. 8 weeks).
SDS 3115 Introduction to Computing For Statistical Sciences
with Professor Bill Katsianos
This course is a first course in computing using statistical software. Topics covered include loading and modifying data, writing functions, and iteration. Emphasis will be placed on data cleaning and data visualization. Students will be able to solve practical problems using software. No previous experience with statistical software is required.
SDS 4480 Topics in Statistics (AI Under the Statistical Microscope: Generalizations, Robustness, and Surprise)
with Professor Apratim Dey
Artificial Intelligence (AI) is rapidly transforming science, technology and society. Today, AI systems can predict the structures of proteins, improve weather forecasts, generate images and text, recommend what we watch and read, assist with hiring and education, and increasingly influence decisions that affect people's lives. Their successes have been remarkable, yet they remain black boxes in many important ways. In this course, we will put modern AI under a statistical microscope. We will study methods to investigate and identify what the AI systems are learning, and as it turns out, many of such ideas have fundamentally statistical roots. We will demonstrate how such statistical ideas provide insights into questions of generalizability, reliability and robustness of AI models, and also discuss surprising deviations from them. Some familiarity with Python may be useful but not essential.
SDS 4481-01 Special Topics in Statistics (Elements of Statistical Thinking (with Big or Small Data)
with Professor Xuming He
Statistical thinking is an essential skill in today's data-driven world. Building on students' prior coursework in statistics and data science, this course explores some of the fundamental statistical principles that underpin modern data science. Through real-world examples, students will examine how statistical reasoning informs every stage of the data lifecycle—from data collection and experimental design to modeling, inference, prediction, and decision-making. In-depth class discussions and carefully designed assignments encourage students to integrate concepts learned across previous courses and develop a deeper understanding of statistical models and methods. By the end of the course, students will be able to synthesize their statistical knowledge and skills, critically assess the strengths and limitations of statistical and machine learning methods, and develop a more mature perspective on the role of statistical thinking in solving complex, real-world problems. In addition, the course will include discussion on data ethics and data quality. The course is organized around the following core elements of statistical thinking:
- Sample vs. population: We explore the role of randomness and experimental design in data analysis, examine what determines the margin of error, why probability-based sampling is important for valid inference.
- Statistical significance: We examine the strengths and limitations of hypothesis testing and p-values, discuss why statistical significance should be interpreted with care, and explore how to move beyond statistical significance in practical decision-making.
- Observational study vs. controlled experiments: We know that correlation is not causation, but we will go further by examining how data collection schemes influence the questions we can answer and the validity of the conclusions we draw from data.
- Purposeful model building: Traditional statistical models prioritize simplicity and interpretability, while large language models have transformed machine learning and AI. We examine the role of models, their capabilities and limitations, and how to use them wisely.
- Integration of data and belief: Data are essential, but integrating multiple data sources with prior belief and domain expertise can substantially improve analysis and decision making. We discuss the principles and challenges of such integration.
New Graduate Course Offerings
SDS 5320 Foundations for Causal Inference
with Professor Chao Cheng
This course offers a comprehensive introduction to the concepts, methods, and applications of causal inference in the social and biomedical sciences. Topics include the potential outcomes framework, assumptions for causal identification, and a range of statistical approaches for estimating various causal effects in both experimental and observational studies, such as propensity score methods, doubly robust estimators, g-computation methods, instrumental variables estimation, and double machine learning estimators. Students will learn to critically evaluate study designs, assess the plausibility of assumptions, and interpret results in context. The course emphasizes both statistical foundations and practical applications, with hands-on exercises in R for implementing methods and analyzing real-world datasets.
SDS 5440 Mathematics Foundations of Big Data
with Professor Todd Kuffner
Mathematical foundations of data science. Core topics include: Probability in high dimensions; curses and blessings of dimensionality; concentration of measure; matrix concentration inequalities. Essentials of random matrix theory. Randomized numerical linear algebra. Data clustering. Depending on time and interests, additional topics will be chosen from: Compressive sensing; efficient acquisition of data; sparsity; low-rank matrix recovery. Divide, conquer and combine methods. Elements of topological data analysis; point cloud; Cech complex; persistent homology. Selected aspects of high- dimensional computational geometry and dimension reduction; embeddings; Johnson-Lindenstrauss; sketching; random projections. Diffusion maps; manifold learning; intrinsic geometry of massive data sets. Optimization and stochastic gradient descent. Random graphs and complex networks. Combinatorial group testing. Prior familiarity with analysis, topology, and geometry is strongly recommended. A willingness to learn new mathematics as needed is essential.
SDS 5480 Topics in Statistics (AI Under the Statistical Microscope: Generalizations, Robustness, and Surprise)
with Professor Apratim Dey
Artificial Intelligence (AI) is rapidly transforming science, technology and society. Today, AI systems can predict the structures of proteins, improve weather forecasts, generate images and text, recommend what we watch and read, assist with hiring and education, and increasingly influence decisions that affect people's lives. Their successes have been remarkable, yet they remain black boxes in many important ways. In this course, we will put modern AI under a statistical microscope. We will study methods to investigate and identify what the AI systems are learning, and as it turns out, many of such ideas have fundamentally statistical roots. We will demonstrate how such statistical ideas provide insights into questions of generalizability, reliability and robustness of AI models, and also discuss surprising deviations from them. Some familiarity with Python may be useful but not essential.
SDS 5800-01 Topics in Statistics (Modern High-Dimensional Statistics and Machine Learning)
with Professor Mengxin Yu
This course provides an in-depth introduction to modern high-dimensional statistics and machine learning, with an emphasis on the recent theory, methods, and algorithms underlying modern data science. Topics include nonparametric regression, high-dimensional sparse regression, sure independence screening, supervised learning, deep learning/AI models (including neural networks, diffusion models, and transformers), uncertainty quantification for black-box model/AI predictions (predictive inference), factor models and covariance learning. Additional topics may include applications across various areas of modern machine learning and artificial intelligence. Throughout the course, the theoretical foundations, practical applicability, and limitations of these methods will be explored through statistical analysis, applications to a variety of real-world datasets, and hands-on implementation using R and Python.
SDS 5971 Topics in Statistics (Environmental Statistics)
with Professor Bo Li
This course explores statistical methods tailored for analyzing complex environmental data, such as those from climatology, ecology, and public health. Students will learn key techniques including time series modeling, spatial data modeling, generalized linear regression, and Bayesian methods. Real-world data examples will be integrated throughout to enhance understanding.
