| Autor: |
Kovalsky, Kristián, Palamas, George |
| Přispěvatelé: |
Shaghaghi, Navid, Lamberti, Fabrizio, Beams, Brian, Shariatmadari, Reza, Amer, Ahmed |
| Jazyk: |
angličtina |
| Rok vydání: |
2021 |
| Předmět: |
|
| Zdroj: |
Kovalsky, K & Palamas, G 2021, Neuroevolution vs Reinforcement Learning for Training Non Player Characters in Games : The Case of a Self Driving Car . in N Shaghaghi, F Lamberti, B Beams, R Shariatmadari & A Amer (eds), Intelligent Technologies for Interactive Entertainment-12th EAI International Conference, INTETAIN 2020, Proceedings . Springer, Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering, vol. 377, pp. 191-206, 12th EAI International Conference on Intelligent Technologies for Interactive Entertainment, INTETAIN 2020, Virtual, Online, 12/12/2020 . https://doi.org/10.1007/978-3-030-76426-5_13 |
| DOI: |
10.1007/978-3-030-76426-5_13 |
| Popis: |
The aim of this project is to compare two popular machine learning methods, a non-gradient-based algorithm such as neuro-evolution with a gradient-based reinforcement learning on an irregular task of training a car to self-drive around 3D circuits with varying complexity. A series of 3D circuits with a physics based car model were modeled using the Unity game engine. The data collected during evaluation show that neuro-evolution converges faster to a solution when compared to the reinforcement learning approach. However, when the reinforcement learning approach is allowed to train for long enough, it outperforms the neuro-evolution in terms of car speed and lap times achieved by the trained model of the car. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
