Integrating LLMs and Decision Transformers for Language Grounded Generative Quality-Diversity
This work addresses the challenge of customizing and generating diverse policies in reinforcement learning and control domains, offering a novel approach for applications like robot navigation, though it is incremental in combining existing techniques.
The paper tackles the problem of generating customizable and diverse trajectories in Quality-Diversity optimization by integrating Large Language Models (LLMs) and Decision Transformers, enabling users to specify arbitrary behavior descriptors and high-level textual prompts to shape trajectories, with experimental validation on a simulated robot navigation benchmark.
Quality-Diversity is a branch of stochastic optimization that is often applied to problems from the Reinforcement Learning and control domains in order to construct repertoires of well-performing policies/skills that exhibit diversity with respect to a behavior space. Such archives are usually composed of a finite number of reactive agents which are each associated to a unique behavior descriptor, and instantiating behavior descriptors outside of that coarsely discretized space is not straight-forward. While a few recent works suggest solutions to that issue, the trajectory that is generated is not easily customizable beyond the specification of a target behavior descriptor. We propose to jointly solve those problems in environments where semantic information about static scene elements is available by leveraging a Large Language Model to augment the repertoire with natural language descriptions of trajectories, and training a policy conditioned on those descriptions. Thus, our method allows a user to not only specify an arbitrary target behavior descriptor, but also provide the model with a high-level textual prompt to shape the generated trajectory. We also propose an LLM-based approach to evaluating the performance of such generative agents. Furthermore, we develop a benchmark based on simulated robot navigation in a 2d maze that we use for experimental validation.