Front-propagation Algorithm: Explainable AI Technique for Extracting Linear Function Approximations from Neural Networks.
Javier ViañaPublished in: CoRR (2024)
Keyphrases
- neural network
- k means
- improved algorithm
- similarity measure
- linear complexity
- pattern recognition
- artificial intelligence
- cost function
- dynamic programming
- times faster
- detection algorithm
- linear approximation
- piecewise linear
- machine learning
- recognition algorithm
- closed form
- optimization algorithm
- computational complexity
- expert systems
- linear programming
- convergence rate
- objective function
- optimal solution
- transfer function
- preprocessing
- computational cost
- worst case
- segmentation method
- high accuracy
- simulated annealing
- segmentation algorithm
- linear systems
- function approximators
- expectation maximization
- computationally tractable
- input data