Learning Graphical Models
Stochastic Segmentation Networks: Modelling Spatially Correlated Aleatoric Uncertainty
In medical imaging, for example, experts will often disagree about the exact location of object boundaries.
Incorporating Interpretable Output Constraints in Bayesian Neural Networks
Work done while at Harvard University.
Adversarially-learned Inference via an Ensemble of Discrete Undirected Graphical Models
Undirected graphical models are compact representations of joint probability distributions over random variables. To solve inference tasks of interest, graphical models of arbitrary topology can be trained using empirical risk minimization.
Contrastive Active Inference
In contrast, reinforcement learning requires human-designed rewards to accomplish any desired outcome.
Supplementary material: High-recall causal discovery for autocorrelated time series with latent confounders
It allows for both latent confounders and selection variables, although in this paper we assume the absence of selection variables.
Bayesian Bellman Operators Matthew Fellows Kristian Hartikainen Shimon Whiteson Department of Computer Science University of Oxford
We demonstrate that BootDQNprior+'s lagged target parameters, which are essential to its performance, arise from applying approximate inference to the BBO posterior.
Variational inference via Wasserstein gradient flows
Below, we proceed to describe our contributions in greater detail. Our framework builds upon the seminal work of Jordan et al. [1998], which introduced the celebrated