Limitations of benchmark sets and landscape features for algorithm selection and performance prediction.
Benjamin LacroixJohn A. W. McCallPublished in: GECCO (Companion) (2019)
Keyphrases
- computational cost
- selection algorithm
- detection algorithm
- prediction algorithm
- prediction error
- times faster
- optimization algorithm
- dynamic programming
- learning algorithm
- recognition algorithm
- theoretical analysis
- search space
- experimental evaluation
- input data
- k means
- optimal solution
- image features
- extracted features
- expectation maximization
- feature extraction
- segmentation algorithm
- search algorithm
- convergence rate
- data sets
- matching process
- improved algorithm
- similarity measure
- video sequences
- key features
- tree structure
- prediction accuracy
- graph cuts
- linear programming
- feature set