Purification Of Contaminated Convolutional Neural Networks Via Robust Recovery: An Approach with Theoretical Guarantee in One-Hidden-Layer Case.
Hanxiao LuZeyu HuangRen WangPublished in: CoRR (2024)
Keyphrases
- hidden layer
- convolutional neural networks
- theoretical guarantees
- back propagation
- learning rate
- feed forward
- radial basis function
- neural network
- artificial neural networks
- recurrent neural networks
- activation function
- feedforward neural networks
- multilayer perceptron
- neural nets
- feed forward neural networks
- learning algorithm
- data fusion
- input output
- linear combination
- particle swarm optimization
- small number