The Best of Many Worlds: Scheduling Machine Learning Inference on CPU-GPU Integrated Architectures.
Giorgos VasiliadisRafail TsirbasSotiris IoannidisPublished in: IPDPS Workshops (2022)
Keyphrases
- knowledge base
- heterogeneous computing
- machine learning
- knowledge acquisition
- graphics processing units
- compute intensive
- gpu implementation
- graphics processors
- scheduling problem
- statistical inference
- data mining
- scheduling algorithm
- grid computing
- high performance computing
- bayesian networks
- machine learning algorithms
- machine learning methods
- text classification
- feature selection
- parallel computing
- single instruction multiple data
- massively parallel
- resource constraints
- explanation based learning
- probabilistic inference
- round robin
- real time
- natural language processing
- pattern recognition
- probabilistic model
- parallel processors
- real time transaction processing
- inference process
- belief networks
- inductive logic programming
- learning tasks
- text mining
- information extraction
- active learning
- computer science
- decision trees