Beginner · Reinforcement Learning
Greedy policy
Visual diagram · (in preparation) · Math · (in preparation) · Worked example · 3 difficulty levels.
TL;DR. A reinforcement learning policy that always selects the action expected to yield the highest reward.
Technical Definition
A reinforcement learning policy that always selects the action expected to yield the highest reward.
How it works
In reinforcement learning, a greedy policy is a strategy where an agent always chooses the action that currently appears to be the best. This means it picks the action with the highest immediate expected return, without considering potential future consequences or exploring other options that might be better in the long run. It's a simple strategy but can sometimes get stuck in suboptimal solutions.
Related Concepts
- Reinforcement Learning — A paradigm where an agent learns to make decisions by receiving rewards or penalties from its environment through trial and error.
- Policy — In reinforcement learning, the strategy an agent follows to choose actions given a state.