Verifying Attention Robustness of Deep Neural Networks Against Semantic Perturbations.
Satoshi MunakataCaterina UrbanHaruki YokoyamaKoji YamamotoKazuki MunakataPublished in: NFM (2023)
Keyphrases
- neural network
- neural network model
- artificial neural networks
- semantic information
- semantic annotation
- domain independent
- focus of attention
- back propagation
- semantic web
- multi layer
- feed forward
- computational efficiency
- context dependent
- semantic relationships
- semantic representation
- natural language understanding
- semantically rich
- domain specific
- fuzzy logic
- natural language
- pattern recognition
- artificial intelligence
- self organizing maps
- higher level
- recurrent neural networks
- training process
- semantic knowledge
- high level
- semantically meaningful
- genetic algorithm