An OpenCL™ Deep Learning Accelerator on Arria 10.
Utku AydonatShane O'ConnellDavor CapalijaAndrew C. LingGordon R. ChiuPublished in: FPGA (2017)
Keyphrases
- deep learning
- message passing interface
- unsupervised learning
- unsupervised feature learning
- parallel implementation
- field programmable gate array
- machine learning
- shared memory
- restricted boltzmann machine
- mental models
- weakly supervised
- training set
- multiscale
- natural images
- markov random field
- parallel computing
- co occurrence
- deep architectures