Personlig profil
Profil informationstekst
Ehsan Yousefzadeh-Asl-Miandoab is a Postdoctoral Researcher at the IT University of Copenhagen (ITU), Denmark. His work lies at the intersection of computer systems, architecture, and emerging applications, with a particular focus on optimizing hardware efficiency for deep learning training tasks.
He earned his PhD at ITU, where his research tackled the challenge of GPU underutilization in deep learning training. His doctoral thesis explored both analytical studies and system designs aimed at improving GPU resource usage and training efficiency.
Earlier, during his master’s studies in computer architecture, Ehsan investigated GPU energy efficiency at the microarchitectural level, proposing new designs for on-chip memories to reduce power consumption while maintaining performance.
His broader research interests include GPUs, parallel computing systems, energy-efficient architectures, and the use of machine learning and deep learning techniques to guide efficient system design.
Emneord
- Resource-aware Machine Learning
- Developing mechanisms
- System efficient
- Workloads
- AI sustainable
Fingeraftryk
- 1 Lignende profiler
Samarbejde og topforskningsområder i de sidste fem år
Publikation
-
Profiling and Monitoring Deep Learning Training Tasks
Yousefzadeh-Asl-Miandoab, E., Robroek, T. T. & Tözün, P., 2023, I: EuroMLSys '23. s. 18–25Publikation: Artikel i tidsskrift og konference artikel i tidsskrift › Konferenceartikel › Forskning › peer review
Åben adgang -
An Analysis of Collocation on GPUs for Deep Learning Training
Robroek, T. T., Yousefzadeh-Asl-Miandoab, E. & Tözün, P., 22 apr. 2024, Proceedings of the 4th Workshop on Machine Learning and Systems, EuroMLSys 2024, Athens, Greece, 22 April 2024. Association for Computing Machinery, s. 81-90 10 s.Publikation: Konference artikel i Proceeding eller bog/rapport kapitel › Konferencebidrag i proceedings › Forskning › peer review
Åben adgang -
Data Management and Visualization for Benchmarking Deep Learning Training Systems
Robroek, T. T., Duane, A., Yousefzadeh-Asl-Miandoab, E. & Tözün, P., 2023, Proceedings of the Seventh Workshop on Data Management for End-to-End Machine Learning, DEEM 2023, Seattle, WA, USA, 18 June 2023. Association for Computing Machinery, s. 1:1-1:5 5 s.Publikation: Konference artikel i Proceeding eller bog/rapport kapitel › Konferencebidrag i proceedings › Forskning › peer review
Åben adgang -
Collocation and Memory Estimation for Efficient GPU Resource Management in Deep Learning Training
Yousefzadeh-Asl-Miandoab, E., 2025, IT-Universitetet i København. 141 s.Publikation: Afhandlinger › Ph.d.-afhandling
Fil -
DP-Morph: Improving the Privacy-Utility-Performance Trade-off for Differentially Private OCT Segmentation
Parsarad, S., Yousefzadeh-Asl-Miandoab, E., Kafieh, R., Tözün, P., Ciorba, F. M. & Wagner, I., 13 okt. 2025, Proceedings of the 18th ACM Workshop on Artificial Intelligence and Security. Association for Computing Machinery, s. 264-275 12 s.Publikation: Konference artikel i Proceeding eller bog/rapport kapitel › Konferencebidrag i proceedings › Forskning › peer review
Åben adgang
-
RAD: Extremely Parallel and Incredibly Diverse Data Processing on Many Heterogeneous Cores
Tözün, P. (PI), Yousefzadeh-Asl-Miandoab, E. (CoI) & Robroek, T. T. (CoI)
01/04/2021 → 30/04/2025
Projekter: Projekt › Forskning
-
RAD+: Resource-Aware Data Science
Tözün, P. (PI), Rosero, P. (CoI), Nielsen , N. K. (CoI), Tøttrup, J. V. (CoI), Bayer, R. (CoI), Duane, A. (CoI), Hvass Jørgensen, J. (CoI), Osterhammel, J. M. (CoI), Sørensen, P. K. (Admin) & Yousefzadeh-Asl-Miandoab, E. (Samarbejdspartner)
01/12/2021 → 31/03/2025
Projekter: Projekt › Forskning
-
DEEP: Deep Learning Resource-Efficient GPU Orchestrator
Tözün, P. (PI) & Yousefzadeh-Asl-Miandoab, E. (Samarbejdspartner)
FONDO NAZIONALE SVIZZERO PER LA RICERCA SCIENTIFICA
01/08/2025 → 31/07/2029
Projekter: Projekt › Forskning
Aktiviteter
- 1 Besøger en ekstern, akademisk institution
-
University of Basel
Yousefzadeh-Asl-Miandoab, E. (Gæsteforsker)
1 sep. 2023 → 1 dec. 2023Aktivitet: Besøg på en ekstern institution - typer › Besøger en ekstern, akademisk institution