ConceptAI & ML
Data parallelism in training
A training parallelization strategy where multiple inputs are run in parallel across machines, requiring coordinated weight updates via a parameter server.
Know first
Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.
Lecturers
1Connections
3The lectures that define it or assume it, and the concepts next to it.