conftrace_

model parallel

4 papers

Explore in graph

Co-occurring keywords · papers mentioning both

batch training (1) persistent computation (1) reduced precision (1) stale synchronous parallel (1) large batch training (1) transformer model (1) translation quality (1) graphics processing unit (1) memory footprint (1) recurrent neural network (1)

Papers

Distributed Proximal Gradient Algorithm for Partially Asynchronous Computer Clusters JMLR 2018 Scaling Neural Machine Translation EMNLP 2018 On Convergence of Model Parallel Proximal Gradient Algorithm for Stale Synchronous Parallel System AISTATS 2016 Persistent RNNs: Stashing Recurrent Weights On-Chip ICML 2016