Artificial Intelligence
深度强化学习
Deep Reinforcement LearningA machine learning paradigm that enables agents to learn optimal behaviors by interacting with an environment through trial and error, maximizing cumulative rewards. / 深度强化学习是一种机器学习范式，智能体通过与环境的交互（试错）来学习最优行为，以最大化累积奖励。
- Knowledge type
- 数学理论
- First proposed or discovered
- Proposer
- Turkey Farm member
- Creator
- 农场主
- Last updated
- 2026-10-23 07:01:41
Sub-knowledge points
Heterogeneous Graph Reinforcement Learning / 数学理论Multi-agent Reinforcement Learning / 数学理论Markov Decision Processes (MDP)Q-LearningPolicy GradientsActor-Critic methods
Higher-order knowledge points
Main parameters and performance indicators
environment: The simulated or real-world setting the agent interacts withagent: The learner that makes decisionspolicy: The strategy mapping states to actionsreward: The scalar feedback signal guiding the learning process
Open problems
sample efficiencyexploration-exploitation trade-offstability in continuous control
Related literature
ranked by community starsState-enhanced surrogate environment and multi-time-scale deep reinforcement learning for robotic grinding optimisationOpen source
Related products and links
No item recorded yet
Creation and modification trail
1 recordsRecord title: Deep Reinforcement Learning.
Node update feedback
review required