Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/156774
Full metadata record
DC FieldValueLanguage
dc.contributor.authorRibli, Vincenten_US
dc.date.accessioned2022-04-23T12:57:21Z-
dc.date.available2022-04-23T12:57:21Z-
dc.date.issued2022-
dc.identifier.citationRibli, V. (2022). Elementwise overparameterisation for single and multi-task learning. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/156774en_US
dc.identifier.urihttps://hdl.handle.net/10356/156774-
dc.description.abstractIn the field of autonomous vehicles, computer vision is used to solve multiple tasks such as semantic segmentation and object tracking. This can be challenging as the tasks need to be done at a high performance within a given latency threshold. Furthermore, multiple tasks need to be solved simultaneously within the given constraints. To solve such an issue, methods such as overparameterisation, along with multi-task learning, have been proposed. This paper proposes a novel overparameterisation technique, along with a few training tricks, which achieves empirically superior performance compared to existing approaches. These ideas are firstly tested on the CIFAR-100 dataset, which is a single-task problem performing image classification. The ideas are further tested on a multi-task setting using the NYUv2 dataset, performing semantic segmentation, depth estimation and surface normals estimation simultaneously. The results of experimentation have shown promising results through the novel overparameterisation approach, and it is hoped that this overparameterisation technique can generalise well to other architectures and datasets as a simple, yet effective approach to improve performance of deep learning models.en_US
dc.language.isoenen_US
dc.publisherNanyang Technological Universityen_US
dc.subjectEngineering::Computer science and engineeringen_US
dc.titleElementwise overparameterisation for single and multi-task learningen_US
dc.typeFinal Year Project (FYP)en_US
dc.contributor.supervisorSinno Jialin Panen_US
dc.contributor.schoolSchool of Computer Science and Engineeringen_US
dc.description.degreeBachelor of Science in Data Science and Artificial Intelligenceen_US
dc.contributor.supervisoremailsinnopan@ntu.edu.sgen_US
item.grantfulltextembargo_restricted_20221021-
item.fulltextWith Fulltext-
Appears in Collections:SCSE Student Reports (FYP/IA/PA/PI)
Files in This Item:
File Description SizeFormat 
FYP Report_Vincent Ribli.pdf
  Until 2022-10-21
6 MBAdobe PDFUnder embargo until Oct 21, 2022

Page view(s)

47
Updated on Jul 2, 2022

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.