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: CVPR (2020)
Keyphrases
- human activities
- motion capture data
- neural network
- human subjects
- data sets
- knowledge base
- empirical data
- neural network model
- prior knowledge
- conceptual model
- computational model
- input data
- expert knowledge
- measured data
- experimental data
- raw data
- predictive model
- human experts
- simulation data
- data mining
- test data
- domain experts
- probability distribution
- data analysis
- e learning
- genetic algorithm
- data collection
- data mining techniques
- fuzzy logic
- attitudes toward
- learning process
- authoring tool
- artificial neural networks
- data structure
- metadata