LLMs have a lot of hype around them these days. Let’s demystify how they work and see how we can put them in context for data science use. As data scientists, we want to make sure our results are inspectable, reliable, reproducible, and replicable. We already have many tools to help us in this front. However, LLMs provide a new challenge; we may not always be given the same results back from a query. This means trying to work out areas where LLMs excel in, and use those behaviours in our data science artifacts. This talk will introduce you to LLms, the Ellmer, and Chatlas packages for R and Python, and how they can be integrated into a Shiny to create an AI-powered dashboard. We’ll see how we can leverage the tasks LLMs are good at to better our data science products.

To join this seminar virtually, please request Zoom connection details from ea@stat.ubc.ca. 

The relative risk (RR) offers interpretation and comparison advantages over the Odds Ratio (OR) used in logistic regression. However, its direct estimation in high-dimensional settings is challenging. Common approaches, such as penalized log-binomial and Poisson regression, are built on parameters that are variationally dependent, while newer, variation-independent models have been limited by estimators not designed for high-dimensional or sparse data.

To address this, this project built on previous penalized RR models to implement a faster penalized estimator for the variation-independent relative risk model. The contributions include an efficient implementation in C++, the use of an Adaptive Step Size FISTA algorithm for robust optimization, and a comprehensive evaluation of different penalization strategies and model specifications. Through simulation studies, the proposed estimator is shown to be a robust tool for high-dimensional analysis. It demonstrates better predictive accuracy and the ability to identify relevant predictors in sparse scenarios correctly.

To join this seminar virtually, please request Zoom connection details from ea@stat.ubc.ca. 

Estimating the parameters of disease transmission models is an important component in analyzing disease outbreaks and inferring transmission networks. The introduction of genetic data into disease transmission models has enabled more detailed inference, particularly through phylogenetic trees derived from the genetic data. Existing approaches often rely on a single phylogenetic tree to subset transmission trees from a set of possible transmission trees inferred from epidemiological data. However, such methods typically do not account for the uncertainty inherent in phylogenetic reconstruction. This thesis introduces a Sequential Monte Carlo-Expectation Maximization (SMC-EM) framework that explicitly incorporates uncertainty in transmission and phylogenetic trees. We treat these trees as latent variables and use observed genetic sequences, sampling times, and epidemiological data to inform the model. Our method constructs transmission and phylogenetic trees sequentially, conditioned on infection times, and updates parameter estimates iteratively via a variant of the EM algorithm. We evaluate the performance of the proposed method through extensive simulation studies and demonstrate its applicability using a real-world outbreak dataset. The results indicate that the SMC-EM approach provides improved parameter estimates while effectively capturing the uncertainty in latent tree structures. 

To join this seminar virtually, please request Zoom connection details from ea@stat.ubc.ca. 

There are many non-Gaussian time series models available in the literature. Copula-based time series models are particularly relevant as they can handle serial tail dependence or the clustering of extreme observations. To date, mainly copula-based Markov time series models that extend the autoregressive time series model have been studied and applied. In this talk, I will consider non-Markovian copula-based time series models that can be viewed as an extension of Gaussian autoregressive moving average (ARMA) models. I derive distributional properties and discuss conditions for stationarity, as well as the asymptotic properties of the maximum-likelihood estimators. Finally, the probabilistic forecasting performance is evaluated.

To join this seminar virtually, please request Zoom connection details from ea@stat.ubc.ca. 

DNA methylation is a key epigenetic mechanism governing gene regulation and cellular identity. Advances in high-throughput sequencing technologies have enabled detailed investigation of methylation landscapes across single cells and complex tissue mixtures. However, the sparsity and noise inherent in single-cell data, as well as the signal distortion in enrichment-based platforms, pose major analytical challenges. This thesis presents two novel statistical frameworks to address these limitations and advance the computational toolkit for DNA methylation analysis.

The first contribution is vmrseq, a probabilistic method and software for detecting variably methylated regions from single-cell bisulfite sequencing data. vmrseq integrates a smoothing-based strategy for candidate region identification with hidden Markov modeling to account for spatial correlation and technical noise. Through extensive benchmarking on synthetic and experimental datasets, vmrseq demonstrates improved precision and biological relevance in identifying methylation heterogeneity, supporting downstream analyses such as unsupervised clustering and cell-type-specific marker discovery.

The second contribution is decemedip, a hierarchical Bayesian model and software for cell type deconvolution of enrichment-based methylation data such as MeDIP-seq. By leveraging reference panels derived from alternative platforms and modeling the complex relationship between methylation levels, CpG density, and read counts, decemedip enables accurate estimation of cell type proportions with uncertainty quantification. Its performance is validated through simulations, cross-platform comparisons, and real-world applications involving patient-derived xenografts and circulating cell-free DNA from cancer cohorts.

Together, these methods address critical gaps in the analysis of high-throughput DNA methylation data, enabling robust detection of epigenetic heterogeneity across biological contexts. The associated open-source software implementations provide practical tools for future epigenomic research and potential clinical applications.

To join this seminar virtually, please request Zoom connection details from ea@stat.ubc.ca. 

It is difficult to anticipate the myriad challenges that a predictive model will encounter once deployed. Common practice entails a reactive, cyclical approach: model deployment, data mining, and retraining. We instead develop a proactive longtail discovery process by imagining additional data during training. In particular, we develop general model-based longtail signals, including a differentiable, single forward pass formulation of epistemic uncertainty that does not impact model parameters or predictive performance but can flag rare or hard inputs. We leverage these signals as guidance to generate additional training data from a latent diffusion model in a process we call Longtail Guidance (LTG). Crucially, we can perform LTG without retraining the diffusion model or the predictive model, and we do not need to expose the predictive model to intermediate diffusion states. Data generated by LTG exhibit semantically meaningful variation, yield significant generalization improvements on numerous image classification benchmarks, and can be analyzed by a VLM to proactively discover, textually explain, and address conceptual gaps in a deployed predictive model.

Bio

David Hayden leads Perception AI Research at Cruise, where he focuses on generative and world models, foundation model alignment and guidance, longtail robustness, uncertainty quantification, and synthetic data. He has consulted on machine learning and computer vision for diverse industries including pharmaceuticals, retail, and competitive sports. His work has shipped to hundreds of driverless cars, ran live in stadiums of 40,000 people, supported seed and Series A rounds, and is published in top conferences and journals including ICML, CVPR, Neurips, and Nature. He previously founded Essistive Technologies, where he developed and licensed discreet note-taking tech for individuals with limited vision. David received a PhD at MIT working on interpretable machine learning and computer vision, with emphasis on behavior analysis, multi-object tracking, Bayesian nonparametrics for time-series, distributions on manifolds, and uncertainty to guide decision making. 

To join this seminar virtually, please request Zoom connection details from ea@stat.ubc.ca. 

Variational inference with stochastic gradients, commonly called black-box variational inference (BBVI) or stochastic gradient variational inference, is the workhorse of probabilistic inference in the large data, large model regime. For a decade, however, the computational properties of VI have largely been unknown. For instance, under what conditions is BBVI guaranteed to converge, and is it provably efficient? In this talk, I will present recent theoretical results on VI in the form of quantitative non-asymptotic convergence guarantees for obtaining a variational posterior. Following this, I will demonstrate the usefulness of the theoretical framework by investigating the theoretical properties of various design choices and algorithmic modifications, such as parametrizations of variational approximation, variance-reduced gradient estimators such as sticking-the-landing, structured variational families, and beyond.

To join this seminar virtually, please request Zoom connection details from ea@stat.ubc.ca. 

To join this seminar virtually: Please request Zoom connection details from ea@stat.ubc.ca 

Abstract: The first passage time (FPT) is a useful tool in stochastic modeling of many biological, physical, social and economic processes evolving with time. It refers to the time when a random process first passes a threshold, e.g., when the population of an endangered species reaches a certain critical level, or when the number of infected individuals with a disease reaches a limit. Other examples include the survival time of a cancer patient, failure time of a mechanical system, and default time of a business, etc.

We study the boundary crossing problem for jump-diffusion processes over a discontinuous boundary and provide a complete characterization on the FPT distributions. We derive new formulas for piecewise linear boundary crossing probabilities and density of Brownian motion with general random jumps. These formulas can be used to approximate the boundary crossing distributions for general nonlinear boundaries. The method can be extended to more general diffusion processes such as geometric Brownian motion and Ornstein-Uhlenbeck processes with jumps. The numerical computation can be done by Monte Carlo integration which is straightforward and easy to implement. Some numerical examples are presented for illustration.

To join this seminar virtually: Please request Zoom connection details from ea@stat.ubc.ca

Presentation 1

Time: 11:00am – 11:30am

Speaker: Xihan (Henry) Qian, UBC Statistics MSc student

Title:  Modeling Diatom Dynamics and Environmental Drivers Using Functional Regression and Rank-Based Model Selection

Abstract: In this project, we analyze the relationship between diatom concentrations and environmental drivers in the Salish Sea using functional data analysis (FDA). Daily measurements of solar radiation, wind speed, air temperature, and diatom levels from 2007 to 2024 are treated as smooth functions over time. To capture the delayed effects of environmental variables on diatom dynamics, we apply a historical functional linear model and estimate a time-varying coefficient surface using finite element basis functions defined over a triangular domain. Smoothness is enforced using directional roughness penalties. Two model selection strategies are compared: one based on the Bayesian Information Criterion (BIC), and another prioritizing predictive performance using mean squared error and rank correlation. We show how the choice of tuning parameters affects predictive accuracy and highlight patterns in the estimated effects of environmental variables on diatom levels.

Presentation 2

Time: 11:30am – 12:00pm

Speaker: Joey Hotz, UBC Statistics MSc student

Title: The Rocky Road Toward Effective Vanilla Bayesian Optimization in High-Dimensional Search Spaces

Abstract: Bayesian optimization (BayesOpt) is a well-established statistical methodology for efficiently finding the true optimum value of a black-box function. A common concern with Bayesian optimization is the "Curse of Dimensionality", as these methods often struggle for input spaces with many parameters unless the algorithm is adjusted accordingly. Despite the prevalence of these challenges, a recently published paper empirically demonstrated that under certain specifications for the surrogate model, the sole adjustment required to make Bayesian optimization effective for higher-dimensional problems is to simply scale the prior distribution for the model based on the dimensionality of the search space. In this presentation, we discuss the background, methodology, and findings of the aforementioned paper. Additionally, we significantly broaden the scope of their simulation study to a wider class of statistical models to evaluate the robustness of their stated result.

To join this seminar virtually: Please request Zoom connection details from ea@stat.ubc.ca

Abstract: Statistics education stands at a critical juncture as we navigate the intersection of traditional statistical theory, modern computational approaches, and emerging AI technologies. This talk examines how statisticians can reimagine curricula by embracing computation as foundational elements rather than afterthoughts. While traditional statistics education has prioritized theoretical frameworks and applications, computation has emerged as the backbone of contemporary data analysis—from data acquisition and wrangling to visualization, modeling, and communication. Now, AI tools are further transforming this landscape, creating both opportunities and challenges for statistics and data science educators. The presentation will outline a forward-looking curriculum model for introductory courses that balances statistical thinking, data science methods, and explicit computational instruction