Interpretation of Neural Networks is Susceptible to Universal Adversarial Perturbations.
Haniyeh Ehsani OskouieFarzan FarniaPublished in: ICASSP (2023)
Keyphrases
- neural network
- image interpretation
- artificial neural networks
- pattern recognition
- multi agent
- feed forward
- semantic interpretation
- neural nets
- neural network model
- feedforward neural networks
- fuzzy logic
- recurrent neural networks
- multilayer perceptron
- fuzzy systems
- multi layer
- neural learning
- adaptive resonance theory
- data sets
- soft computing
- back propagation
- expert systems
- training data
- website
- image processing
- decision making
- social networks
- artificial intelligence
- genetic algorithm