Events

A major component of my research focuses on quantifying the health impacts of climate-related hazards and modelling population dynamics using detailed and large datasets, on scales ranging from small-area to multi-country, for which Bayesian modelling can afford numerous advantages. In this seminar, I will highlight some of my major research efforts on climate, health, and equity, including several recent and ongoing studies in the United States, Chile, and the Philippines. Focus topics will include natality and mortality disparities, studies of the association of health-relevant outcomes with heat stress, tropical cyclones, and wildfires, and ongoing work on climate change and health attribution.

Characterizing how neural organoids respond to electrical stimulation is essential for understanding their computational and adaptive properties. In this study, we develop a Bayesian hierarchical framework to analyze spike activity recorded from organoids using multi-electrode array (MEA) platforms under varying stimulation intensities. Spike counts are modeled as stochastic processes, allowing us to quantify stimulation effects while accounting for electrode-level variability, repeated trials, and parameter uncertainty.

Using Bayesian generalized linear modeling and posterior inference, we estimate stimulation-dependent changes in firing rates and obtain uncertainty quantification. The hierarchical structure enables partial pooling stimulation levels, improving statistical efficiency and robustness.

Our results show systematic modulation of spike activity with increasing stimulation intensity, alongside substantial heterogeneity across recording sites. This probabilistic framework provides a principled and flexible approach for analyzing organoid electrophysiology data and supports inference in novel biological neural systems. 

The increasing availability of large-scale geospatial and spatiotemporal data presents new opportunities and challenges for statistical modeling in environmental, technological, medical, and other complex areas, which increasingly rely on massive multivariate spatiotemporal datasets. Yet, Bayesian learning for such problems remains severely limited by computational bottlenecks and the lack of flexible modeling tools. Modern applications require methods that are adaptive and effective, but still computationally efficient, scalable to massive datasets, and capable of delivering reliable automated inference with principled uncertainty quantification and (possibly) minimal experienced human intervention. Classical Bayesian approaches, although theoretically appealing and offering rich inferential frameworks, often become computationally infeasible in data-rich environments, especially when confronted with massive datasets or dynamic, high-dimensional dependence structures. Existing approaches often fail to scale, leaving a gap between the theoretical richness of Bayesian inference and its practical deployment in data-rich applications. This thesis develops Bayesian transfer learning methodologies to address these challenges, enabling efficient information propagation and scalable inference across large spatial and spatiotemporal domains, providing a unified framework that merges distributional theory for matrix-variate models with computational innovations in Bayesian predictive stacking. Through extensive simulation experiments and data applications to global and satellite monitoring of vegetation indices, sea surface temperature, and land-atmospheric climate composition, the thesis also demonstrates the potential of Bayesian transfer learning to redefine spatial and spatiotemporal multivariate modeling, providing flexible, computationally efficient solutions that open the way for scalable, automated, and truly modern tools for geospatial learning in data-rich environments.

Modern scientific studies routinely record multiple predictors alongside various correlated outcomes, often of mixed types, such as continuous measurements and binary disease indicators. Analysing such data outcome-by-outcome can ignore residual dependence, distort uncertainty quantification, and reduce power in high dimensions. This talk introduces Bayesian tools for learning multivariate structure and heterogeneous effects in high dimensions, with scalable computation and theoretical guarantees.

I will first present mixed-mSSL, a joint regression framework for mixed-type multivariate responses built on latent Gaussian augmentation. By combining spike-and-slab LASSO priors on regression effects with sparse graphical priors on the residual precision matrix, mixed-mSSL simultaneously selects predictors and learns an outcome-dependence network. A scalable Monte Carlo ECM algorithm enables MAP estimation, and we establish posterior contraction rates for both the coefficient matrix and the precision matrix and support recovery guarantees under diverging outcome-dimensions. mixed-mSSL demonstrates excellent finite-sample properties, using extensive simulation studies and applications spanning medicine to ecology.

Next, I move beyond constant effects to settings where covariate impacts may vary with context. I will discuss sparseVCBART, which places BART ensembles on varying-coefficient functions while inducing two-way sparsity: selecting relevant covariates and identifying which modifiers drive effect heterogeneity. As a natural extension, I will also outline ongoing work on multivariate BART for multiple correlated outcomes, allowing outcome-adaptive tree structure while borrowing strength via a shared residual covariance.

In many scientific applications, uncertainty of estimates from an upstream analysis needs to be propagated into a downstream analysis without feedback. Cutting feedback, or cut-Bayes, achieves this by constructing a cut-posterior distribution that prevents backward information flow. However, sampling-based implementations of cutting feedback, like nested MCMC, are computationally intensive, while existing variational inference (VI) approaches require two approximations and access to upstream data. We propose NeVI-Cut, a neural network-based variational inference method for cutting feedback. NeVI-Cut directly uses upstream samples without requiring access to upstream data, avoiding extra approximation error. We employ normalizing flows, neural network-based generative models, to flexibly model the downstream conditional distribution. We provide theoretical guarantees for NeVI-Cut to approximate any cut-posterior. A triply stochastic algorithm implements the method. Simulation studies and two real-world analysis show that NeVI-Cut achieves substantial computational gains over sampling-based cutting feedback methods and is more accurate than parametric variational cut approaches.