LLMs for HVAC Control
First page
Paper summary
Microsoft applies LLMs to industrial control tasks (HVAC for buildings), comparing against RL baselines.
Ask this paper
01
Demonstration selection: Develops a recipe for selecting demonstrations and generating high-performing prompts for industrial control tasks.
02
GPT-4 ≈ RL: GPT-4 performs comparably to specialized RL methods on HVAC control, despite being a general-purpose model.
03
Lower technical debt: Uses dramatically fewer samples and avoids the operational complexity of training and maintaining a dedicated RL policy.
04
Practical implication: Suggests LLMs can substitute for RL in many control tasks where sample efficiency and maintenance matter more than peak performance.