An enhancement of the neuroevolutionary algorithm applied in Super Mario Bros Game
By: Eryck Luis R. Arceo and Mark Jason S. Escopete
Language: English . . 2016Description: Undergraduate Thesis: (BSCS major in Computer Science) - Pamantasan ng Lungsod ng Maynila, 2016Content type: text Media type: unmediated Carrier type: volumeGenre/Form: academic writingDDC classification: . LOC classification: QA76.9 Ar3 2016| Item type | Current location | Home library | Collection | Call number | Status | Date due | Barcode | Item holds |
|---|---|---|---|---|---|---|---|---|
| Archival materials | PLM | PLM Archives | Filipiniana-Thesis | QA76.9 Ar3 2016 (Browse shelf) | Available | FT6068 |
Browsing PLM Shelves , Shelving location: Archives , Collection code: Filipiniana-Thesis Close shelf browser
ABSTRACT: NeuroEvolutionary is an algorithm that uses evolution algorithm to train artificial neutral networks to be presented for the people to have an enhanced working Artificial Intelligence for solving a stage in Super Mario Bros. The research involves determining the problems of the existing algorithm. This has been done by examining past related studies and literatures and pinpointing the common problems encountered. This also involves proving that the NeuroRevolutionary algorithm is the best algorithm to be used in applications with network topology by comparing different types of evolutionary algorithms. The research also discusses the methods and procedures it underwent. A survey was conducted to outline the performance of the test, simulate and calibrate to attain the best possible result. The anticipated outcome of the research is the improved initialization, minimizing network and reducing the output which was presented during the simulation and the data gathering. This has been done by comparing and interpreting the results of the existing and enhanced algorithm. Upon the completion of the research, the researchers provided recommendations which may be highly recommended for other aspiring researchers to improve.
5
ABSTRACT: NeuroEvolutionary is an algorithm that uses evolution algorithm to train artificial neutral networks to be presented for the people to have an enhanced working Artificial Intelligence for solving a stage in Super Mario Bros. The research involves determining the problems of the existing algorithm. This has been done by examining past related studies and literatures and pinpointing the common problems encountered. This also involves proving that the NeuroRevolutionary algorithm is the best algorithm to be used in applications with network topology by comparing different types of evolutionary algorithms. The research also discusses the methods and procedures it underwent. A survey was conducted to outline the performance of the test, simulate and calibrate to attain the best possible result. The anticipated outcome of the research is the improved initialization, minimizing network and reducing the output which was presented during the simulation and the data gathering. This has been done by comparing and interpreting the results of the existing and enhanced algorithm. Upon the completion of the research, the researchers provided recommendations which may be highly recommended for other aspiring researchers to improve.
Filipiniana
5

There are no comments for this item.