AAEBERT: Debiasing BERT-based Hate Speech Detection Models via Adversarial Learning.
Ebuka OkpalaLong ChengNicodemus MbwamboFeng LuoPublished in: ICMLA (2022)
Keyphrases
- learning process
- learning systems
- accurate models
- learning tasks
- hidden variables
- structured prediction
- knowledge acquisition
- online learning
- hidden markov models
- complex systems
- active learning
- prior knowledge
- learning problems
- e learning
- discriminative learning
- learned models
- learning rules
- inductive inference
- word processing
- neural nets
- automatic detection
- mobile learning
- false positives
- detection algorithm
- anomaly detection
- model selection