Artificial Intelligence
异构图强化学习
Heterogeneous Graph Reinforcement LearningA machine learning paradigm that integrates reinforcement learning with graph neural networks to optimize decision-making processes within complex, multi-relational data structures. / 一种将强化学习与图神经网络相结合的机器学习范式，用于在复杂的多关系数据结构中优化决策过程。
- Knowledge type
- 数学理论
- First proposed or discovered
- Proposer
- Turkey Farm member
- Creator
- 农场主
- Last updated
- 2026-10-23 06:01:32
Sub-knowledge points
Graph representation learningMarkov Decision Process (MDP) formulationPolicy optimization algorithms
Higher-order knowledge points
Main parameters and performance indicators
state: The current graph structure and node featuresaction: Dispatching or scheduling decisionsreward: System efficiency or buffer constraint satisfactionpolicy: The learned strategy mapping states to actions
Open problems
Scalability to very large-scale logistics networksGeneralization across different types of unrelated parallel machinesHandling dynamic and stochastic buffer constraints in real-time
Related literature
ranked by community starsHub-centric heterogeneous graph reinforcement learning for integrated automated guided vehicle dispatching and unrelated parallel machine scheduling with finite-buffer constraintsOpen source
Related products and links
No item recorded yet
Creation and modification trail
1 recordsRecord title: Heterogeneous Graph Reinforcement Learning.
Node update feedback
review required