원문정보
초록
영어
Reinforcement learning can be applied to a wide variety of problems. However, the fundamental limitation of reinforcement learning is that it is difficult to derive an answer within a given time because the problems in the real world are too complex. Then, with the development of neural network technology, research on deep reinforcement learning that combines deep learning with reinforcement learning is receiving lots of attention. In this paper, two types of neural networks are combined with reinforcement learning and their characteristics were compared and analyzed with existing value-based reinforcement learning algorithms. Two types of neural networks are FNN and CNN, and existing reinforcement learning algorithms are SARSA and Qlearning.
목차
1. INTRODUCTION
2. Cart-pole system
3. Value-Based Reinforcement Learning
3.1. Traditional reinforcement learning
3.2. Reinforcement learning with Neural Network
4. EMPIRICAL RESULTS AND OBSERVATION
4.1. RL Reward of cart-pole system
4.2. Pole Oscillation Angle of cart-pole system
5. CONCLUSION
ACKNOWLEDGEMENT
REFERENCES