Q-learning
Sign in to save'''Q-learning''' is a reinforcement learning algorithm that trains an agent to assign values to its possible actions based on its current state, without requiring a model of the environment (model-free). It can handle problems with stochastic transitions and rewards without requiring adaptations.
Article
19 sectionsContents
- Reinforcement learning
- Algorithm
- Influence of variables
- Learning rate
- Discount factor
- Initial conditions (''Q''<sub>0</sub>)
- Implementation
- Function approximation
- Quantization
- History
- Variants
- Deep Q-learning
- Double Q-learning
- Others
- Multi-agent learning
- Limitations
- See also
- References
- External links
'''Q-learning' is a reinforcement learning algorithm that trains an agent to assign values to its possible actions based on its current state, without requiring a model of the environment (model-free). It can handle problems with stochastic transitions and rewards without requiring adaptations.
For example, in a grid maze, an agent learns to reach an exit worth 10 points. At a junction, Q-learning might assign a higher value to moving right than left if right gets to the exit faster, improving this choice by trying both directions over time.