Language-guided Skill Learning with Temporal Variational Inference

arXiv:2402.16354v1 Announce Type: new
Abstract: We present an algorithm for skill discovery from expert demonstrations. The algorithm first utilizes Large Language Models (LLMs) to propose an initial segmentation of the trajectories. Following that, a hierarchical variational inference framework incorporates the LLM-generated segmentation information to discover reusable skills by merging trajectory segments. To further control the trade-off between compression and reusability, we introduce a novel auxiliary objective based on the Minimum Description Length principle that helps guide this skill discovery process. …

abstract algorithm arxiv cs.ai cs.cl cs.lg discovery expert framework generated hierarchical inference information language language models large language large language models llm llms merging segmentation skills temporal the algorithm trajectory type