Daegun Yoon

Publication Activity (10 Years)

Years Active: 2020-2024
Publications (10 Years): 15

Top Topics

Space Partitioning

Top Venues

J. Supercomput.

Publications

Daegun Yoon, Sangyoon Oh
Preserving Near-Optimal Gradient Sparsification Cost for Scalable Distributed Deep Learning. CoRR (2024)
Daegun Yoon, Sangyoon Oh
MiCRO: Near-Zero Cost Gradient Sparsification for Scaling and Accelerating Distributed DNN Training. CoRR (2023)
Daegun Yoon, Sangyoon Oh
DEFT: Exploiting Gradient Norm Difference between Model Layers for Scalable Gradient Sparsification. ICPP (2023)
Daegun Yoon, Sangyoon Oh
MiCRO: Near-Zero Cost Gradient Sparsification for Scaling and Accelerating Distributed DNN Training. HiPC (2023)
Daegun Yoon, Minjoong Jeong, Sangyoon Oh
WAVE: designing a heuristics-based three-way breadth-first search on GPUs. J. Supercomput. 79 (6) (2023)
Daegun Yoon, Minjoong Jeong, Sangyoon Oh
SAGE: toward on-the-fly gradient compression ratio scaling. J. Supercomput. 79 (10) (2023)
Seungjun Lee, Miri Yu, Daegun Yoon, Sangyoon Oh
Can hierarchical client clustering mitigate the data heterogeneity effect in federated learning? IPDPS Workshops (2023)
Daegun Yoon, Sangyoon Oh
DEFT: Exploiting Gradient Norm Difference between Model Layers for Scalable Gradient Sparsification. CoRR (2023)
Juwon Park, Daegun Yoon, Sangho Yeo, Sangyoon Oh
AMBLE: Adjusting mini-batch and local epoch for federated learning with heterogeneous devices. J. Parallel Distributed Comput. 170 (2022)
Daegun Yoon, Sangyoon Oh
Empirical Analysis on Top-k Gradient Sparsification for Distributed Deep Learning in a Supercomputing Environment. CoRR (2022)
Daegun Yoon, Sangyoon Oh
SURF: Direction-Optimizing Breadth-First Search Using Workload State on GPUs. Sensors 22 (13) (2022)
Daegun Yoon, Zhetao Li, Sangyoon Oh
Balanced content space partitioning for pub/sub: a study on impact of varying partitioning granularity. J. Supercomput. 77 (12) (2021)
Seungjun Lee, Daegun Yoon, Sangho Yeo, Sangyoon Oh
Mitigating Cold Start Problem in Serverless Computing with Function Fusion. Sensors 21 (24) (2021)
Daegun Yoon, Gyudong Park, Sangyoon Oh
Exploring a system architecture of content-based publish/subscribe system for efficient on-the-fly data dissemination. Concurr. Comput. Pract. Exp. 33 (18) (2021)
Daegun Yoon, Gyudong Park, Sangyoon Oh
CPartition: a Correlation-Based Space Partitioning for Content-Based Publish/Subscribe Systems with Skewed Workload. BigComp (2020)