On-FPGA Training with Ultra Memory Reduction: A Low-Precision Tensor Method.
Kaiqi ZhangCole HawkinsXiyuan ZhangCong HaoZheng ZhangPublished in: CoRR (2021)
Keyphrases
- high speed
- cost function
- reduction method
- training algorithm
- high order
- support vector machine
- detection method
- objective function
- high accuracy
- training phase
- training process
- support vector machine svm
- higher order
- probabilistic model
- dynamic programming
- segmentation algorithm
- significant improvement
- pairwise
- computational complexity