Does Compressing Activations Help Model Parallel Training?

Song Bian Dacheng Li Hongyi Wang Eric P. Xing Shivaram Venkataraman

Published in: MLSys (2024)

Keyphrases

computational model
information retrieval
cost function
training data
theoretical analysis
parameter estimation
machine learning
knowledge base
prior knowledge
probabilistic model
probability distribution
test set
simulation model
formal model