Introduction and Exemplars of Uncertainty Decomposition

TL;DR

This paper introduces the concept of uncertainty decomposition in machine learning, explaining its importance for high-stakes applications and illustrating various decomposition methods through examples like neural networks and Gaussian processes.

Contribution

It provides a clear introduction to the types of uncertainty and exemplifies decomposition techniques across different models, enhancing understanding of uncertainty quantification.

Findings

Clarifies the notion of uncertainty decomposition

Provides exemplars across multiple models

Connects to broader topics in uncertainty analysis

Abstract

Uncertainty plays a crucial role in the machine learning field. Both model trustworthiness and performance require the understanding of uncertainty, especially for models used in high-stake applications where errors can cause cataclysmic consequences, such as medical diagnosis and autonomous driving. Accordingly, uncertainty decomposition and quantification have attracted more and more attention in recent years. This short report aims to demystify the notion of uncertainty decomposition through an introduction to two types of uncertainty and several decomposition exemplars, including maximum likelihood estimation, Gaussian processes, deep neural network, and ensemble learning. In the end, cross connections to other topics in this seminar and two conclusions are provided.