Certifying Neural Network Robustness to Random Input Noise from Samples.
Brendon G. AndersonSomayeh SojoudiPublished in: CoRR (2020)
Keyphrases
- neural network
- input data
- desired output
- greater robustness
- statistically independent
- noise level
- data sets
- sampled data
- noise immunity
- back propagation
- noise reduction
- general regression neural networks
- random noise
- image noise
- genetic algorithm
- median filter
- noise model
- noisy data
- random samples
- missing data
- bp neural network
- feed forward
- neural network model
- fault diagnosis
- artificial neural networks
- geometric attacks
- training data