Maximum-likelihood estimators in physics-informed neural networks for high-dimensional inverse problems.
Gabriel S. GusmãoAndrew J. MedfordPublished in: Comput. Chem. Eng. (2024)
Keyphrases
- inverse problems
- maximum likelihood
- high dimensional
- neural network
- image reconstruction
- global optimization
- parameter space
- convex optimization
- maximum likelihood estimator
- optimization methods
- optimization problems
- pattern recognition
- dimensionality reduction
- emission tomography
- partial differential equations
- genetic algorithm
- early vision
- expectation maximization
- maximum a posteriori
- computationally expensive
- smoothness constraint
- feature space
- super resolution
- mutual information
- high resolution
- evolutionary algorithm
- image analysis
- object recognition
- multiscale