Multi-head Knowledge Distillation for Model Compression.
Huan WangSuhas LohitMichael JonesYun FuPublished in: CoRR (2020)
Keyphrases
- computational model
- real time
- mathematical model
- probabilistic model
- prior knowledge
- statistical model
- theoretical analysis
- expert systems
- formal model
- conceptual model
- cost function
- domain knowledge
- probability distribution
- knowledge management
- knowledge sharing
- experimental data
- compression algorithm
- formal representation
- additional knowledge
- input data
- knowledge representation
- multiresolution
- high level
- decision trees
- information systems
- data sets