Repository navigation
Augmented Lagrangian constraint
Introduces comprehensive support for constrained optimization using the Augmented Lagrangian method, transforming the optimizer from a simple distance minimization tool into a flexible constraint-based system.
Major Features
- Implements Augmented Lagrangian (ConstrainedFitnessAL) for constrained CMA-ES optimization
- Supports both hard constraints and soft constraints
- Flexible constraint definition using GeoGebra expressions
- Configurable tolerance and weights for fine-grained control
- Batch evolution processing for performance improvement