We study the problem of uncertainty quantification for time series prediction, with the goal of
providing easy-to-use algorithms with formal guarantees. The algorithms we present build …

Methods for reasoning under uncertainty are a key building block of accurate and reliable
machine learning systems. Bayesian methods provide a general framework to quantify …

Confidence calibration–the problem of predicting probability estimates representative of the
true correctness likelihood–is important for classification models in many applications. We …

Modern neural networks have recently been found to be poorly calibrated, primarily in the
direction of over-confidence. Methods like entropy penalty and temperature smoothing …

Optimal decision making requires that classifiers produce uncertainty estimates consistent
with their empirical accuracy. However, deep neural networks are often under-or over …

Label shift refers to the phenomenon where the prior class probability p (y) changes
between the training and test distributions, while the conditional probability p (x| y) stays …

Many applications of classification methods not only require high accuracy but also reliable
estimation of predictive uncertainty. However, while many current classification frameworks …

Deep neural networks (DNNs) have shown to perform very well on large scale object
recognition problems and lead to widespread use for real-world applications, including …

Accurate probabilistic predictions can be characterized by two properties {—} calibration and
sharpness. However, standard maximum likelihood training yields models that are poorly …

Reliable uncertainty estimates are an important tool for helping autonomous agents or
human decision makers understand and lever-age predictive models. However, existing …