Path-Weights and Layer-Wise Relevance Propagation for Explainability of ANNs with fMRI Data.
José Diogo Marques dos SantosJosé Paulo Marques dos SantosPublished in: LOD (2) (2023)
Keyphrases
- artificial neural networks
- relative importance
- hidden neurons
- using artificial neural networks
- feedforward artificial neural networks
- output layer
- weighted sum
- hidden nodes
- connection weights
- information retrieval
- multi layer
- application layer
- genetic algorithm ga
- feed forward
- shortest path
- artificial neural
- linear combination
- feed forward artificial neural networks
- synaptic weights
- feed forward neural networks
- feature weighting
- activation function
- hidden layer
- pairwise
- input variables
- feedforward neural networks
- term weights
- neural nets
- test collection
- back propagation
- weight function
- improved harmony search algorithm
- learning algorithm
- web pages
- application of artificial neural networks
- evolutionary algorithm
- wave propagation
- edge weights
- generalization ability
- neural network