University of Twente Student Theses
Systematic Comparison Of Auto encoders in an Inudstrial Application
Klimkevičius, Kipras (2024) Systematic Comparison Of Auto encoders in an Inudstrial Application.
|
PDF
657kB |
| Abstract: | Anomaly detection is a field of AI that has significant benefits to many fields, from medical tests to industrial cases. How- ever, anomaly detection models could become expensive to train, due to it being very expensive and difficult to gather enough data to train supervised models. Unsupervised mod- els do not have this issue, since they can be trained on an imbalanced dataset, which as the name of the issue implies, is mostly the case in anomaly detection. Here we system- atically compare Auto Encoder (AE) based approaches to unsupervised anomaly detection. For this purpose, we use an industrial dataset and find that Variational AEs perform more consistently and with better results than Convolutional AEs. |
| Item Type: | Essay (Bachelor) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 54 computer science |
| Programme: | Computer Science BSc (56964) |
| Link to this item: | https://purl.utwente.nl/essays/100875 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page