There is Strength in Numbers: Avoiding the Hypothesis-Only Bias in Natural Language Inference via Ensemble Adversarial Training.
Joe StaceyPasquale MinerviniHaim DubossarskySebastian RiedelTim RocktäschelPublished in: CoRR (2020)
Keyphrases
- natural language
- training set
- structured prediction
- training process
- training data
- ensemble learning
- neural network
- semantic interpretation
- test set
- ensemble methods
- natural language interface
- natural language understanding
- semantic analysis
- training examples
- decision trees
- learning algorithm
- random forest
- conditional random fields
- random forests
- training algorithm
- knowledge representation
- inference engine
- hypothesis space
- artificial neural networks
- machine learning