A vision for edge AI
| dc.contributor.advisor | Chen, Jian-Jia | |
| dc.contributor.author | Yayla, Mikail | |
| dc.contributor.referee | Teich, Jürgen | |
| dc.date.accepted | 2024-02-28 | |
| dc.date.accessioned | 2024-04-10T11:03:23Z | |
| dc.date.available | 2024-04-10T11:03:23Z | |
| dc.date.issued | 2024 | |
| dc.description.abstract | Edge Artificial Intelligence is progressively pervading all aspects of our life. However, to perform complex tasks, a massive amount of matrix multiplications needs to be computed. At the same time, the available hardware resources for computation are highly limited. The pressing need for efficiency serves as the motivation for this dissertation. In this dissertation, we propose a vision for highly-resource constrained future intelligent systems that are comprised of robust Binarized Neural Networks operating with approximate memory and approximate computing units, while being able to be trained on the edge. | en |
| dc.identifier.uri | http://hdl.handle.net/2003/42431 | |
| dc.identifier.uri | http://dx.doi.org/10.17877/DE290R-24267 | |
| dc.language.iso | en | de |
| dc.subject | Embedded system | en |
| dc.subject | Edge AI | en |
| dc.subject | Machine learning | en |
| dc.subject | HW/SW codedesign | en |
| dc.subject | Non-volatile memory | en |
| dc.subject | Error tolerance | en |
| dc.subject | Efficient training | en |
| dc.subject.ddc | 004 | |
| dc.subject.rswk | Künstliche Intelligenz | de |
| dc.subject.rswk | Eingebettetes System | de |
| dc.subject.rswk | Maschinelles Lernen | de |
| dc.subject.rswk | Fehlertoleranz | de |
| dc.title | A vision for edge AI | en |
| dc.title.alternative | robust binarized neural networks on emerging resource-constrained hardware | en |
| dc.type | Text | de |
| dc.type.publicationtype | PhDThesis | de |
| dcterms.accessRights | open access | |
| eldorado.secondarypublication | false | de |