Reinforcement learning (RL), particularly in sparse reward settings, often requires prohibitively large numbers of interactions with the environment, thereby limiting its applicability to complex problems. To address this, several prior approaches have used natural language to guide the agent's exploration. However, these approaches typically operate on structured representations of the environment, and/or assume some structure in the natural language commands. In this work, we propose a model that directly maps pixels to rewards, given a free-form natural language description of the task, which can then be used for policy learning. Our experiments on the Meta-World robot manipulation domain show that language-based rewards significantly improves the sample efficiency of policy learning, both in sparse and dense reward settings.
翻译:强化学习(RL),特别是在微薄的奖赏环境中,往往要求大量与环境互动,从而限制其适用于复杂问题。为了解决这个问题,以前的若干方法已经使用自然语言指导代理人的勘探,但是,这些方法通常以环境结构化的形式运作,并(或)在自然语言指令中采用某种结构。在这项工作中,我们提出了一个模型,根据该任务的自由形式自然语言描述,直接绘制奖赏的像素图,然后用于政策学习。我们在Meta-World机器人操纵域的实验显示,基于语言的奖励极大地提高了政策学习的抽样效率,无论是在稀少还是密集的奖赏环境中都是如此。