Towards implementation of residual-feedback GMDH neural network on parallel GPU memory guided by a regression curve.
Ricardo BritoSimon FongKyungeun ChoWei SongRaymond K. WongSabah MohammedJinan FiaidhiPublished in: J. Supercomput. (2016)
Keyphrases
- neural network
- parallel implementation
- graphics processing units
- cluster of workstations
- graphics processors
- parallel processing
- parallel hardware
- parallel computation
- group method of data handling
- single instruction multiple data
- parallel programming
- auto associative
- graphics cards
- parallel computing
- general purpose
- parallel algorithm
- real time
- associative memory
- parallel computers
- computer architecture
- graphics hardware
- floating point
- processing elements
- multi threaded
- artificial neural networks
- model selection
- regression model
- memory bandwidth
- endpoints
- gpu implementation
- genetic algorithm
- neural network model
- massively parallel
- parallel architectures
- parallel architecture
- feedback loop
- hardware implementation
- compute unified device architecture
- memory requirements
- efficient implementation