Adaptive co-evolution using reinforcement learning
The field of co-evolution is focused on evolving agents based on the fitness of another evolving agent; in this way, agents are expected to learn from each other. This research aims to develop agents, which would adapt to the skill level of an individual human. Particularly, this study investigates...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | text |
| Published: |
Animo Repository
2007
|
| Subjects: | |
| Online Access: | https://animorepository.dlsu.edu.ph/faculty_research/3229 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | De La Salle University |
| Summary: | The field of co-evolution is focused on evolving agents based on the fitness of another evolving agent; in this way, agents are expected to learn from each other. This research aims to develop agents, which would adapt to the skill level of an individual human. Particularly, this study investigates and discusses the behavior of reinforcement learning (RL) as an algorithm for co-evolution. A modified snake game was used to test and evaluate co-evolutionary behavior exhibited by the RL algorithm against genetic algorithm and best response learning algorithm |
|---|