Abstract Concept Bottleneck Models (CBMs) tackle the opacity of neural architectures by
constructing and explaining their predictions using a set of high-level concepts. A special …

Deep generative models like variational autoencoders (VAEs) are widely used for density
estimation and dimensionality reduction, but infer latent representations via amortized …

Amortized Bayesian inference trains neural networks to solve stochastic inference problems
using model simulations, thereby making it possible to rapidly perform Bayesian inference …

Cognitive scientists believe that adaptable intelligent agents like humans perform spatial
reasoning tasks by learned causal mental simulation. The problem of learning these …

Cognitive scientists believe adaptable intelligent agents like humans perform reasoning
through learned causal mental simulations of agents and environments. The problem of …

We propose an importance sampling method for tractable and efficient estimation of
counterfactual expressions in general settings, named Exogenous Matching. By minimizing …

The philosopher Immanuel Kant formulated the Categorical Imperative as the supreme
principle of all moral beings to do what is “universally good”[9, 10]. A logical consequence is …

This work presents novel techniques for approaching missing data using generative models.
The main focus of these techniques is on leveraging the latent spaces of generative models …

Guidelines: 1. Review the state-of-the-art in sum-product networks in the context of density
modeling, summarize its advantages and disadvantages. Illustrate their benefits on simple …

Výzkum a škálovatelnost výzkumu v oblasti strojového učení se urychlily přechodem od
ručního vytváření příznaků k automatické extrakci příznaků. Použití datového formátu JSON …