Glossary term
Glossary term
Agentic Systems
In reinforcement learning, a policy that always chooses the action with the highest expected return.
Created for this library
A routing engine uses a greedy policy as a baseline to compare against a learned RL dispatcher in simulated city networks.
A trading team uses a greedy policy of always taking the highest-scoring action as the baseline for its RL agent.
A logistics team uses a greedy nearest-neighbor policy as a baseline so leadership can quantify the lift from a learned model.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License