Neural Networks Are More Productive Teachers Than Human Raters: Active Mixup for Data-Efficient Knowledge Distillation from a Blackbox Model.
Dongdong WangYandong LiLiqiang WangBoqing GongPublished in: CoRR (2020)
Keyphrases
- prior knowledge
- experimental data
- neural network
- data sets
- probability distribution
- expert knowledge
- data mining techniques
- raw data
- data analysis
- data collection
- database
- input data
- simulation data
- data structure
- missing information
- training data
- statistical methods
- human subjects
- empirical data
- measured data
- test data
- network structure
- predictive model
- probabilistic model
- knowledge sharing
- conceptual model
- human experts
- domain experts
- high level