Optimal Input-Dependent Edge-Cloud Partitioning for RNN Inference.
Daniele Jahier PagliariRoberta ChiaroYukai ChenEnrico MaciiMassimo PoncinoPublished in: ICECS (2019)
Keyphrases
- dynamic programming
- recurrent neural networks
- cloud computing
- worst case
- edge information
- probabilistic inference
- closed form
- statistically independent
- data sets
- inference process
- computing platform
- user input
- edge detection
- denoising
- special case
- optimal solution
- multiscale
- database systems
- decision trees
- image processing
- databases