Sangyoon Oh

Publication Activity (10 Years)

Years Active: 2023-2024
Publications (10 Years): 11

Top Topics

Heuristic Function

Scalable Distributed

Lossless Coding

Top Venues

J. Supercomput.

IEEE Trans. Serv. Comput.

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)
Miri Yu, Oh-Kyoung Kwon, Sangyoon Oh
Addressing Client Heterogeneity in Synchronous Federated Learning: The CHAFL Approach. ICPADS (2023)
Daegun Yoon, Minjoong Jeong, Sangyoon Oh
SAGE: toward on-the-fly gradient compression ratio scaling. J. Supercomput. 79 (10) (2023)
Sangho Yeo, Minho Bae, Minjoong Jeong, Oh-Kyoung Kwon, Sangyoon Oh
Crossover-SGD: A gossip-based communication in distributed deep learning for alleviating large mini-batch problem and enhancing scalability. Concurr. Comput. Pract. Exp. 35 (15) (2023)
Seungjun Lee, Miri Yu, Daegun Yoon, Sangyoon Oh
Can hierarchical client clustering mitigate the data heterogeneity effect in federated learning? IPDPS Workshops (2023)
Shujuan Tian, Chi Chang, Saiqin Long, Sangyoon Oh, Zhetao Li, Jun Long
User Preference-Based Hierarchical Offloading for Collaborative Cloud-Edge Computing. IEEE Trans. Serv. Comput. 16 (1) (2023)
Daegun Yoon, Sangyoon Oh
DEFT: Exploiting Gradient Norm Difference between Model Layers for Scalable Gradient Sparsification. CoRR (2023)