Learning in coaching: Using neural network to generate advice for Simsoccer coach
Robot Soccer is considered as a platform or a test bed for innovations in research areas such as Artificial Intelligence and Robotics. A recent addition to the RoboCup League is the competition for a special agent called the Coach. Based on the study of current soccer coach agents, learning algorith...
Saved in:
| Main Author: | |
|---|---|
| Format: | text |
| Language: | English |
| Published: |
Animo Repository
2007
|
| Subjects: | |
| Online Access: | https://animorepository.dlsu.edu.ph/etd_masteral/3588 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | De La Salle University |
| Language: | English |
| Summary: | Robot Soccer is considered as a platform or a test bed for innovations in research areas such as Artificial Intelligence and Robotics. A recent addition to the RoboCup League is the competition for a special agent called the Coach. Based on the study of current soccer coach agents, learning algorithm is an important factor for the coach agent. Many classification methods and machine learning algorithms have been suggested in the literature to tackle this problem. In this paper, the neural network rule extraction technique was used to generate advice for Simsoccer Coach through analyzing three fixed opponents behaviors. Furthermore, learning performance of neural network algorithm was compared to C4.5. |
|---|