Hardware-aware Training Techniques for Improving Robustness of Ex-Situ Neural Network Transfer onto Passive TiO2 ReRAM Crossbars.
Philippe DroletRaphaël DawantVictor YonPierre-Antoine MounyMatthieu ValdenaireJavier Arias-ZapataPierre GliechSean U. N. WoodSerge EcoffeyFabien AlibartYann BeilliardDominique DrouinPublished in: CoRR (2023)
Keyphrases
- neural network
- training process
- training algorithm
- feed forward neural networks
- feedforward neural networks
- multi layer perceptron
- artificial neural networks
- back propagation
- train a neural network
- real time
- neural network training
- training patterns
- multi layer
- low cost
- feed forward
- training data
- neural network model
- error back propagation
- hidden layer
- recurrent networks
- fuzzy artmap
- genetic algorithm
- neural network structure
- decision trees
- backpropagation algorithm
- digital circuits
- training phase
- training examples
- neural nets
- massively parallel
- computer systems
- training set
- thin film
- computing systems
- network architecture
- test set
- hardware and software
- self organizing maps