Disentangling Latent Risk Pathways via Bayesian Hypergraph Inference

A batch of new arXiv stat.ML submissions covering methodological advances across Bayesian and variational inference, causal discovery, transfer learning, optimal transport on manifolds, reinforcement learning, and statistical methods for biomedical/clinical data. Highlights include: a Bayesian hypergraph model for interpretable multi-disease pathways in EHRs with scalable variational inference; transfer-learning adaptations for causal forests estimating CATE; identifiability and estimation results for unlabeled finite mixtures using marginal independence; barycentric projections of optimal-­-�