Extreme sparsification of physics-augmented neural networks for interpretable model discovery in mechanics.
Jan Niklas FuhgReese E. JonesNikolaos BouklasPublished in: CoRR (2023)
Keyphrases
- neural network
- mathematical model
- prior knowledge
- cost function
- computational model
- probability distribution
- probabilistic model
- high level
- multi agent
- multi layer
- bayesian networks
- machine learning
- autoregressive
- multiscale
- formal model
- conceptual model
- experimental data
- theoretical framework
- genetic algorithm
- maximum likelihood
- knowledge discovery
- decision making
- data sets
- evolutionary algorithm
- computer science