Neural network based AI for racing games
The usage of neural network-based artificial intelligence in game industry is still minimal, particularly affected by the difficulty and the lengthy training of the neural network itself. Moreover, deciding neural network’s many parameters is very difficult. Yet, it has a lot of potential as a machi...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English |
| Published: |
2010
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/36282 |
| _version_ | 1826114689967325184 |
|---|---|
| author | Dhani Wiryawan Susilo |
| author2 | Narendra Shivaji Chaudhari |
| author_facet | Narendra Shivaji Chaudhari Dhani Wiryawan Susilo |
| author_sort | Dhani Wiryawan Susilo |
| collection | NTU |
| description | The usage of neural network-based artificial intelligence in game industry is still minimal, particularly affected by the difficulty and the lengthy training of the neural network itself. Moreover, deciding neural network’s many parameters is very difficult. Yet, it has a lot of potential as a machine learning technique where it can learn by itself which will help the development of the AI.
This report presents simplified topological evolution method for evolutionary neural network algorithm called NEAT by changing the topology structure to a fully-connected multi layered perceptron (MLP) structure. It also proposes the new mutation function to be used along with the new topological structure. The performance of the new method is then tested using a 2D racing car simulation model in three tracks of increasing difficulty. Our modified evolution method results in simpler neural network topology. Specifically for our experiment, our method results in significantly lesser number of neurons and layers than NEAT. It is further shown that the adaptation maintains the performance of the original algorithm while reducing the evolution time by 17 percent. |
| first_indexed | 2024-10-01T03:43:35Z |
| format | Thesis |
| id | ntu-10356/36282 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T03:43:35Z |
| publishDate | 2010 |
| record_format | dspace |
| spelling | ntu-10356/362822023-03-04T00:34:17Z Neural network based AI for racing games Dhani Wiryawan Susilo Narendra Shivaji Chaudhari School of Computer Engineering Game Lab DRNTU::Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence DRNTU::Engineering::Computer science and engineering::Computer systems organization::Computer-communication networks The usage of neural network-based artificial intelligence in game industry is still minimal, particularly affected by the difficulty and the lengthy training of the neural network itself. Moreover, deciding neural network’s many parameters is very difficult. Yet, it has a lot of potential as a machine learning technique where it can learn by itself which will help the development of the AI. This report presents simplified topological evolution method for evolutionary neural network algorithm called NEAT by changing the topology structure to a fully-connected multi layered perceptron (MLP) structure. It also proposes the new mutation function to be used along with the new topological structure. The performance of the new method is then tested using a 2D racing car simulation model in three tracks of increasing difficulty. Our modified evolution method results in simpler neural network topology. Specifically for our experiment, our method results in significantly lesser number of neurons and layers than NEAT. It is further shown that the adaptation maintains the performance of the original algorithm while reducing the evolution time by 17 percent. Master of Engineering (SCE) 2010-04-30T02:55:24Z 2010-04-30T02:55:24Z 2010 2010 Thesis http://hdl.handle.net/10356/36282 en 91 p. application/pdf |
| spellingShingle | DRNTU::Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence DRNTU::Engineering::Computer science and engineering::Computer systems organization::Computer-communication networks Dhani Wiryawan Susilo Neural network based AI for racing games |
| title | Neural network based AI for racing games |
| title_full | Neural network based AI for racing games |
| title_fullStr | Neural network based AI for racing games |
| title_full_unstemmed | Neural network based AI for racing games |
| title_short | Neural network based AI for racing games |
| title_sort | neural network based ai for racing games |
| topic | DRNTU::Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence DRNTU::Engineering::Computer science and engineering::Computer systems organization::Computer-communication networks |
| url | http://hdl.handle.net/10356/36282 |
| work_keys_str_mv | AT dhaniwiryawansusilo neuralnetworkbasedaiforracinggames |