Ein Angebot des JRZ ISIA
Die Reading Group hat den Charakter eines lockeren Fachseminars mit Vorträgen einer Länge zwischen 30 und 45 Minuten und anschließender Diskussion. In freundlicher Atmosphäre und unter weitestgehender Themenfreiheit werden etwa Forschungsergebnisse, wissenschaftliche oder technologische Überblicksvorträge oder die Aufbereitung einer Forschungsfrage präsentiert und diskutiert. Der Fokus liegt im Austausch, in der Diskussion. Organisatorische Fragen, Anmeldungen zu Vorträgen oder Vortragsvorschläge bitte an Stefan Huber richten.
Termine 2026
May, 27th, 2026 | 15.15 pm | HS 151 | Sebastian Unsin (HS Kempten)
Efficient Machine Learning for Industrial Time Series Data based on Low-Dimensional Representations
Machine Learning for Industrial Time Series Data often suffers from the large input size of such data, particularly when dealing with high-resolution sensors. Deep learning models typically scale poorly with input dimensionality, resulting in resource-intensive architectures that are difficult to train, evaluate, and interpret. While feature extraction methods can mitigate this issue, they tend to discard information critical for many downstream tasks. In this talk, we explore approaches that aim to provide dimensionally reduced representations of such data as a preparation for downstream learning tasks. Our research interest is how these methods can be designed to remain resource-efficient; how domain knowledge can be incorporated to improve interpretability; and to what extent learned representations transfer to other learning tasks. We discuss findings through two industrial case studies: one addressing the scenario of few but very high-resolution sensors, and another involving a large number of low-resolution signals.
Room: HS 151, 15:15pm
If you want to join, please send an email to Stefan Huber.