Explores causal discovery using latent variable models, emphasizing the challenges and solutions in inferring causal relationships from non-Gaussian data.
Explores the integration of machine learning into discrete choice models, emphasizing the importance of theory constraints and hybrid modeling approaches.
Explores the Debiased Whittle likelihood for time series and spatial data, focusing on fitting spectral density to the periodogram for better predictions and parameter estimation.
Explores learning latent models in graphical structures, focusing on scenarios with incomplete samples and introducing the notion of distance among variables.
