Moshpit SGD: Communication-Efficient Decentralized Training on Heterogeneous Unreliable Devices

Aug-16-2025, 05:15:55 GMT–Neural Information Processing Systems

Training deep neural networks on large datasets can often be accelerated by using multiple compute nodes.

all-reduce, international conference, moshpit sgd, (11 more...)

Neural Information Processing Systems

Aug-16-2025, 05:15:55 GMT

Conferences PDF

Country:
- Asia > Russia (0.14)
- Europe
  - Italy > Calabria
    - Catanzaro Province > Catanzaro (0.04)
  - Russia > Central Federal District
    - Moscow Oblast > Moscow (0.04)
- North America
  - Canada > Ontario
    - Toronto (0.14)
  - United States > Massachusetts (0.04)

Industry:
- Energy (0.45)
- Health & Medicine (0.67)
- Information Technology > Services (0.46)

Technology:
- Information Technology
  - Artificial Intelligence
    - Machine Learning > Neural Networks
      - Deep Learning (0.88)
    - Natural Language (1.00)
    - Representation & Reasoning (1.00)
  - Communications (1.00)
  - Data Science (0.92)