Certified machine learning: A posteriori error estimation for physics-informed neural networks.
Birgit HillebrechtBenjamin UngerPublished in: IJCNN (2022)
Keyphrases
- error estimation
- machine learning
- neural network
- model selection
- pattern recognition
- computer science
- error estimates
- support vector machine
- artificial intelligence
- learning algorithm
- machine learning methods
- inductive learning
- machine learning algorithms
- knowledge representation
- natural language processing
- maximum likelihood
- generalization error
- cross validation
- fuzzy logic
- rule extraction
- data mining
- feed forward
- genetic algorithm
- text mining
- decision trees
- artificial neural networks
- lower bound
- explanation based learning
- multi layer
- face recognition
- recurrent neural networks
- learning tasks
- fault diagnosis
- sample size
- back propagation
- active learning
- knowledge acquisition
- text classification
- expert systems
- pairwise
- computational intelligence
- multilayer perceptron
- neural nets
- machine learning approaches
- activation function
- feature selection
- computer vision