Efficient N: M Sparse DNN Training Using Algorithm, Architecture, and Dataflow Co-Design.
Chao FangWei SunAojun ZhouZhongfeng WangPublished in: CoRR (2023)
Keyphrases
- experimental evaluation
- high accuracy
- single pass
- training process
- learning algorithm
- detection algorithm
- optimization algorithm
- computationally efficient
- computational cost
- cost function
- computational complexity
- high efficiency
- worst case
- probabilistic model
- dynamic programming
- significant improvement
- data flow
- k means
- design methodology
- simulated annealing
- preprocessing
- highly efficient
- classification algorithm
- hardware implementation
- floating point
- particle swarm optimization
- software architecture
- matching algorithm
- segmentation algorithm
- bayesian networks
- genetic algorithm
- similarity measure
- np hard
- high dimensional
- optimal solution