Optimiser l'adaptation en ligne d'un module de compréhension de la parole avec un algorithme de bandit contre un adversaire (Adversarial bandit for optimising online active learning of spoken language understanding).
Emmanuel FerreiraAlexandre Reiffers-MassonBassam JabaianFabrice LefèvrePublished in: JEP-TALN-RECITAL (1) (2016)
Keyphrases
- language understanding
- active learning
- random sampling
- natural language understanding
- dialogue management
- online learning
- bandit problems
- semantic interpretation
- language processing
- spoken dialogue systems
- natural language
- general knowledge
- dialogue system
- learning strategies
- machine learning
- semantic analysis
- learning algorithm
- labeled data
- user interaction
- domain specific
- expert systems
- cognitive psychology
- multi agent