When deploying modern machine learning-enabled robotic systems in high-s...
Introduced as a notion of algorithmic fairness, multicalibration has pro...
When deploying machine learning models in high-stakes robotics applicati...
When facing uncertainty, decision-makers want predictions they can trust...
Probabilistic classifiers output confidence scores along with their
pred...
Decision makers often need to rely on imperfect probabilistic forecasts....
Classifiers deployed in high-stakes real-world applications must output
...
Machine learning applications often require calibrated predictions, e.g....
We consider the problem of estimating confidence intervals for the mean ...
Learning generative models for graph-structured data is challenging beca...
We propose a new framework for reasoning about information in complex
sy...
We study the question of how to imitate tasks across domains with
discre...
Partial differential equations (PDEs) are widely used across the physica...
Learning data representations that are transferable and fair with respec...
In high dimensional settings, density estimation algorithms rely crucial...
A variety of learning objectives have been proposed for training latent
...
The variational autoencoder (VAE) is a popular model for density estimat...
Existing Markov Chain Monte Carlo (MCMC) methods are either based on
gen...
It has been previously observed that variational autoencoders tend to ig...
Advances in neural network based classifiers have transformed automatic
...
We propose a new family of optimization criteria for variational
auto-en...
Deep neural networks have been shown to be very successful at learning
f...