Implementation in Groovy of many of the algorithms in Sean Luke's Essentials of Metaheuristics
My goal here is to implement several of the key algorithms in Sean Luke's book Essentials of Metaheuristics in Groovy. My initial goal is to both see what these algorithms look like in Groovy (I'm used to implementing them in Java and, in a few cases, Ruby), and to provide some useful examples for my Evolutionary Computation and Artificial Intelligence course.
Searchers I've implemented:
- A simple implementation of Gradient Ascent for functions of one variable.
- A simple hill-climber
- A steepest ascent hill-climber
- A steepest ascent hill-climber with replacement
Bit-string problems I've implemented:
- OnesMax
- LeadingOnes
- LeadingOnesBlocks
- (Binary) Trap
- HIFF (Hierarchical If-and-only_if)
Real-valued vector problmes I've implemented:
- Sum
- Rastrigin