DP-SGD Without Clipping: The Lipschitz Neural Network Way.
Louis BéthuneThomas MassenaThibaut BoissinYannick PrudentCorentin FriedrichFranck MamaletAurélien BelletMathieu SerrurierDavid VigourouxPublished in: CoRR (2023)
Keyphrases
- neural network
- dynamic programming
- neural network model
- neural network is trained
- artificial neural networks
- pattern recognition
- fuzzy logic
- back propagation
- network architecture
- multilayer perceptron
- stochastic gradient descent
- genetic algorithm
- auto associative
- pointwise
- neural nets
- bp neural network
- training algorithm
- fuzzy neural network
- learning vector quantization
- prediction model
- feed forward
- learning algorithm