Existing agent technologies for complex, multi-step goals either harness symbolic AI to systematically explore potential action spaces or employ reinforcement learning to optimize action trajectories by partitioning tasks into subtasks.

Authors

• Person: Not available Organization: arXiv
  The Synergy of Symbolic and Connectionist AI in LLM-Empowered ...

Sources

• The Synergy of Symbolic and Connectionist AI in LLM-Empowered ... arxiv.org arXiv via serper

Referenced by nodes (2)

• reinforcement learning concept
• Symbolic Artificial Intelligence concept