Nonparametric Inference
Nonparametric inference refers to statistical techniques that use data to infer unknown quantities of interest while making as few assumptions as possible. This can involve working with large and flexible (potentially infinite-dimensional) statistical models, or assuming little about the data-generating process. The flexibility and adaptivity provided by nonparametric techniques is especially valuable in modern statistical problems of the current era of massive datasets and black-box AI.
Berkeley Statistics faculty work on many aspects of nonparametric inference. Current research interests include nonparametric hypothesis testing based on permutations, time-uniform confidence bounds, nonparametric regression, classification and density estimation under smoothness and shape constraints, black-box uncertainty quantification, Bayesian nonparametrics and neural modeling, as well as applications: in particular, to biological research, epidemic forecasting, election auditing, and racial justice in the legal system.
Researchers
statistics, machine learning, semiparametric models, asymptotic theory, hidden Markov models, applications to molecular biology
phase transitions, networks, graphs, graphons, algorithmic game theory, machine learning, applications in cancer immunotherapy, ethical decision-making, climate change, materials science
causal inference, econometrics, experimental design, measurement error, missing data, natural experiments, applications in biomedical and social sciences
high-dimensional statistical learning, statistical computing, computational biology and genomics, precision medicine and health
nonparametric estimation, shape-constrained estimation, high-dimensional statistics, Bayesian and empirical Bayes methods
computer science, artificial intelligence, computational biology, statistics, machine learning, electrical engineering, applied statistics, optimization
machine learning, statistical prediction, variational inference, statistical computing, optimization, sequential inference, causal inference, physical sciences, biology, control and reinforcement learning
language models and diffusion models, deep learning theory, reinforcement learning theory, high-dimensional statistics, quantum algorithms, and uncertainty quantification
uncertainty quantification and inference, inverse problems, nonparametrics, risk assessment, elections, geophysics, astrophysics, cosmology, litigation, health
high-dimensional statistics, nonparametric estimation, distribution-free inference, machine learning, optimization, numerical methods, probabilistic forecasting, computational epidemiology
nonparametric estimation, hypothesis testing, applied probability, statistical learning theory, online learning