paper-conference

Coordinating Distributed Example Orders for Provably Accelerated Training
Recent research on online Gradient Balancing (GraB) has revealed that there exist permutation-based example orderings that are …
A. Feder Cooper, Wentao Guo, Khiem Pham, Tiancheng Yuan, Charlie F. Ruan, Yucheng Lu, Christopher De Sa
PDF Cite Code Poster
Coordinating Distributed Example Orders for Provably Accelerated Training
GraB: Finding Provably Better Data Permutations than Random Reshuffling
Random reshuffling, which randomly permutes the dataset each epoch, is widely adopted in model training because it yields faster …
Yucheng Lu, Wentao Guo, Christopher De Sa
PDF Cite Poster
GraB: Finding Provably Better Data Permutations than Random Reshuffling