Uncertainty-Aware Evaluation of Time-Series Classification for Online Handwriting Recognition With Domain Shift

MCML Authors

David Rügamer

Prof. Dr.

Principal Investigator

Statistics, Data Science and Machine Learning

Bernd Bischl

Prof. Dr.

Director

Statistical Learning and Data Science

Felix Ott

Dr.

* Former Member

→ Group Bernd Bischl
 Statistical Learning and Data Science

Abstract

For many applications, analyzing the uncertainty of a machine learning model is indispensable. While research of uncertainty quantification (UQ) techniques is very advanced for computer vision applications, UQ methods for spatio-temporal data are less studied. In this paper, we focus on models for online handwriting recognition, one particular type of spatio-temporal data. The data is observed from a sensor-enhanced pen with the goal to classify written characters. We conduct a broad evaluation of aleatoric (data) and epistemic (model) UQ based on two prominent techniques for Bayesian inference, Stochastic Weight Averaging-Gaussian (SWAG) and Deep Ensembles. Next to a better understanding of the model, UQ techniques can detect out-of-distribution data and domain shifts when combining right-handed and left-handed writers (an underrepresented group).

inproceedings

---

STRL 2022 @IJCAI-ECAI 2022

Workshop on Spatio-Temporal Reasoning and Learningat the 31st International Joint Conference on Artificial Intelligence and the 25th European Conference on Artificial Intelligence. Vienna, Austria, Jul 23-29, 2022.

---

Authors

A. Klaß • S. M. Lorenz • M. W. Lauer-Schmaltz • D. Rügamer • B. Bischl • C. Mutschler • F. Ott

---

Links

URL

---

Research Area

 A1 | Statistical Foundations & Explainability