Skip to content
conftrace_

large-scale neural network

3 papers

Explore in graph

Co-occurring keywords · papers mentioning both

model parallelism (2) block dropout (1) large-scale training (1) optimization performance (1) independent subnetwork training (1) pipeline parallelism (1) communication compression (1) communication efficiency (1) distributed training (1) quadratic models (1)

Papers

Towards a Better Theoretical Understanding of Independent Subnetwork Training ICML 2024 FedOBD: Opportunistic Block Dropout for Efficiently Training Large-scale Neural Networks through Federated Learning IJCAI 2023 GPipe: Efficient Training of Giant Neural Networks using Pipeline Parallelism NeurIPS 2019