Details
| Originalsprache | Englisch |
|---|---|
| Titel des Sammelwerks | 2023 IEEE/ACM 7th International Workshop on Green And Sustainable Software, GREENS 2023 |
| Seiten | 17-18 |
| Seitenumfang | 2 |
| ISBN (elektronisch) | 9798350312386 |
| Publikationsstatus | Veröffentlicht - 2023 |
Abstract
Deployed machine learning systems often suffer from accuracy degradation over time generated by constant data shifts, also known as concept drift. Therefore, these systems require regular maintenance, in which the machine learning model needs to be adapted to concept drift. The literature presents plenty of model adaptation techniques. The most common technique is periodically executing the whole training pipeline with all the data gathered until a particular point in time, yielding a massive energy footprint. In this paper, we propose a research path that uses concept drift detection and adaptation to enable sustainable AI systems.
ASJC Scopus Sachgebiete
- Informatik (insg.)
- Software
- Energie (insg.)
- Erneuerbare Energien, Nachhaltigkeit und Umwelt
Ziele für nachhaltige Entwicklung
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
2023 IEEE/ACM 7th International Workshop on Green And Sustainable Software, GREENS 2023. 2023. S. 17-18.
Publikation: Beitrag in Buch/Bericht/Sammelwerk/Konferenzband › Aufsatz in Konferenzband › Forschung › Peer-Review
}
TY - GEN
T1 - Retrain AI Systems Responsibly! Use Sustainable Concept Drift Adaptation Techniques
AU - Poenaru-Olaru, Lorena
AU - Sallou, June
AU - Miranda da Cruz, Luis
AU - Rellermeyer, Jan S.
AU - Deursen, Arie van
PY - 2023
Y1 - 2023
N2 - Deployed machine learning systems often suffer from accuracy degradation over time generated by constant data shifts, also known as concept drift. Therefore, these systems require regular maintenance, in which the machine learning model needs to be adapted to concept drift. The literature presents plenty of model adaptation techniques. The most common technique is periodically executing the whole training pipeline with all the data gathered until a particular point in time, yielding a massive energy footprint. In this paper, we propose a research path that uses concept drift detection and adaptation to enable sustainable AI systems.
AB - Deployed machine learning systems often suffer from accuracy degradation over time generated by constant data shifts, also known as concept drift. Therefore, these systems require regular maintenance, in which the machine learning model needs to be adapted to concept drift. The literature presents plenty of model adaptation techniques. The most common technique is periodically executing the whole training pipeline with all the data gathered until a particular point in time, yielding a massive energy footprint. In this paper, we propose a research path that uses concept drift detection and adaptation to enable sustainable AI systems.
KW - concept drift adaptation
KW - sustainable model maintenance
KW - sustainable model retraining
UR - http://www.scopus.com/inward/record.url?scp=85168082338&partnerID=8YFLogxK
U2 - 10.1109/greens59328.2023.00009
DO - 10.1109/greens59328.2023.00009
M3 - Conference contribution
SN - 979-8-3503-1239-3
SP - 17
EP - 18
BT - 2023 IEEE/ACM 7th International Workshop on Green And Sustainable Software, GREENS 2023
ER -