A new arXiv paper surveys how large language models can support deep reinforcement learning in managing resources across IoT, edge, and cloud layers. The work focuses on continuous, context-aware decision-making in environments where constraints shift constantly. It positions LLMs as a complement to established DRL techniques for adaptive computing across the computing continuum.
A revised arXiv paper introduces an experience replay method for deep reinforcement learning that prioritizes diversity when sampling past experiences. The authors argue that conventional uniform and prioritized replay strategies often use stored transitions inefficiently, and their approach aims to improve learning efficiency by selecting a more varied set of experiences. The updated version is cross-listed in the cs.AI and cs.LG categories.
A new study presents a deep reinforcement learning framework that trains multiple drone agents to explore and keep watch over virtual wildfire scenes. Across training, the agents progressively developed consistent and effective surveillance behavior. The research points toward coordinated autonomous aircraft for real-world wildfire response.