Auto-weighted Bayesian Physics-Informed Neural Networks and robust estimations for multitask inverse problems in pore-scale imaging of dissolution.
Sarah PerezPhilippe PoncetPublished in: CoRR (2023)
Keyphrases
- inverse problems
- multi task
- neural network
- image reconstruction
- global optimization
- multitask learning
- gaussian processes
- multi task learning
- convex optimization
- image analysis
- optimization problems
- transfer learning
- optimization methods
- bayesian networks
- image processing
- early vision
- high resolution
- evolutionary algorithm
- computer vision
- cost function
- feature vectors
- feature space
- feature selection
- genetic algorithm