DAE-GP: denoising autoencoder LSTM networks as probabilistic models in estimation of distribution genetic programming.
David WittenbergFranz RothlaufDirk SchweimPublished in: GECCO (2020)
Keyphrases
- genetic programming
- denoising
- probabilistic model
- evolutionary computation
- image denoising
- fitness function
- power law
- evolutionary algorithm
- symbolic regression
- grammar guided genetic programming
- gene expression programming
- total variation
- genetic algorithm
- financial forecasting
- regression problems
- natural images
- graphical models
- image processing
- noisy images
- mixture model
- social networks
- complex networks
- generative model
- recurrent neural networks
- bayesian inference
- language model
- probability density
- denoising methods
- wavelet packet
- gaussian distribution
- denoising algorithm
- expectation maximization
- multiscale
- maximum likelihood estimator
- latent variables
- network structure
- neural network