Model Stacking Performance Comparisons for Lifetime Estimation of CMOS ICs

Master Thesis (2023)
Author(s)

K.A.P. du Buf (TU Delft - Electrical Engineering, Mathematics and Computer Science)

Contributor(s)

N. Yorke-Smith – Mentor (TU Delft - Algorithmics)

Sebastijan Dumancic – Mentor (TU Delft - Algorithmics)

H.N. Kekkonen – Graduation committee member (TU Delft - Statistics)

Faculty
Electrical Engineering, Mathematics and Computer Science
Copyright
© 2023 Koen du Buf
More Info
expand_more
Publication Year
2023
Language
English
Copyright
© 2023 Koen du Buf
Graduation Date
27-06-2023
Awarding Institution
Delft University of Technology
Programme
Electrical Engineering
Faculty
Electrical Engineering, Mathematics and Computer Science
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

Integrated circuits are vital in the modern world. Testing these circuits is often a months long process involving measurements at multiple times during long stress tests. In this work, final measurements from such tests are predicted based on early measurements, potentially reducing the time needed for such tests and giving preliminary results. In addition to this problem, research is done into the benefits of model stacking ensambles, and around the performance impact of using a low bit precision. From our experiments, we observe a significant performance improvement when using model stacking variations. We also find that model stacking retains its performance better than all other tested models when using a lower bit precision.

Files

License info not available