Solving higher-order Lane-Emden-Fowler type equations using physics-informed neural networks: benchmark tests comparing soft and hard constraints.
Hubert BatyPublished in: CoRR (2023)
Keyphrases
- hard constraints
- higher order
- benchmark tests
- neural network
- soft constraints
- cost function
- constraint satisfaction problems
- efficient computation
- constraint satisfaction
- nonlinear equations
- multi objective evolutionary
- constraint violations
- performs competitively
- graph cuts
- search space
- multi criteria
- polynomial equations
- genetic algorithm
- penalty function
- combinatorial optimization
- fuzzy logic
- pairwise
- linear programming
- lower bound