LLM-Guided-Reinforcement-Learning

In our project, we replaced the normal epsilon greedy strategy with the suggestion of LLM to guide the agent towards optimal states, leveraging LLM's knowledge of the states.

The results of our experiments are written in the attached Report.