LARG, Language-based Automatic Reward and Goal Generation
This addresses the scalability issue in reinforcement learning for robotics by reducing reliance on human annotations, though it is incremental as it builds on existing LLM capabilities.
The paper tackles the problem of needing extensive human annotations for text-based robotic manipulation tasks by introducing LARG, a method that uses Large Language Models to automatically generate reward and goal functions from task descriptions, enabling scalable policy training without handcrafted rewards.
Goal-conditioned and Multi-Task Reinforcement Learning (GCRL and MTRL) address numerous problems related to robot learning, including locomotion, navigation, and manipulation scenarios. Recent works focusing on language-defined robotic manipulation tasks have led to the tedious production of massive human annotations to create dataset of textual descriptions associated with trajectories. To leverage reinforcement learning with text-based task descriptions, we need to produce reward functions associated with individual tasks in a scalable manner. In this paper, we leverage recent capabilities of Large Language Models (LLMs) and introduce \larg, Language-based Automatic Reward and Goal Generation, an approach that converts a text-based task description into its corresponding reward and goal-generation functions We evaluate our approach for robotic manipulation and demonstrate its ability to train and execute policies in a scalable manner, without the need for handcrafted reward functions.