Towards Heterogeneous Multi-core Accelerators Exploiting Fine-grained Scheduling of Layer-Fused Deep Neural Networks.
Arne SymonsLinyan MeiSteven CollemanPouya HoushmandSebastian KarlMarian VerhelstPublished in: CoRR (2022)
Keyphrases
- fine grained
- neural network
- coarse grained
- multi layer
- scheduling problem
- access control
- multiple layers
- massively parallel
- single layer
- artificial neural networks
- tightly coupled
- grid environment
- concurrency control
- neural nets
- database
- data fusion
- back propagation
- computing systems
- parallel machines
- scheduling algorithm
- privacy preserving
- privacy policies
- database management systems
- data lineage