Algorithms or Actions? A Study in Large-Scale Reinforcement Learning
| Autor: | Anderson R. Tavares, Luiz Chaimowicz, Leandro Soriano Marcolino, Sivasubramanian Anbalagan |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
0209 industrial biotechnology
Computer science Scale (descriptive set theory) 02 engineering and technology Dilemma 020901 industrial engineering & automation Function approximation Action (philosophy) 0202 electrical engineering electronic engineering information engineering Reinforcement learning 020201 artificial intelligence & image processing State (computer science) Set (psychology) Finite set Algorithm |
| Zdroj: | IJCAI |
| DOI: | 10.24963/ijcai.2018/377 |
| Popis: | Large state and action spaces are very challenging to reinforcement learning. However, in many domains there is a set of algorithms available, which estimate the best action given a state. Hence, agents can either directly learn a performance-maximizing mapping from states to actions, or from states to algorithms. We investigate several aspects of this dilemma, showing sufficient conditions for learning over algorithms to outperform over actions for a finite number of training iterations. We present synthetic experiments to further study such systems. Finally, we propose a function approximation approach, demonstrating the effectiveness of learning over algorithms in real-time strategy games. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|