Investigating Random Variations of the Forward-Forward Algorithm for Training Neural Networks.
Fabio GiampaoloStefano IzzoEdoardo PreziosoFrancesco PiccialliPublished in: IJCNN (2023)
Keyphrases
- neural network
- learning algorithm
- cost function
- randomly generated
- optimization algorithm
- detection algorithm
- training algorithm
- training process
- optimal solution
- feed forward neural networks
- training phase
- np hard
- neural network training
- probabilistic model
- recognition algorithm
- times faster
- expectation maximization
- theoretical analysis
- hopfield neural network
- genetic algorithm
- high accuracy
- fuzzy logic
- dynamic programming
- significant improvement
- search space
- preprocessing
- computational complexity
- linear programming
- back propagation
- worst case
- convergence rate
- recurrent neural networks
- k means
- similarity measure