arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2412.14865cs.LG
收藏

策略分层子空间用于持续离线强化学习

Hierarchical Subspaces of Policies for Continual Offline Reinforcement Learning

Ubisoft La Forge(育碧La Forge);H Company;Inria(法国国家信息与自动化研究所);CNRS(法国国家科学研究中心);Centrale Lille(里尔中央理工学院)

Anthony Kobanda, Rémy Portelas, Odalric-Ambrym Maillard, Ludovic Denoyer

AI总结:

HiSPO通过分层策略子空间实现持续离线强化学习,有效解决导航任务中的遗忘与可扩展性问题。

摘要:

我们考虑了一个持续强化学习设置,其中学习智能体必须在不断适应新任务的同时保留已获得的技能集,重点在于避免遗忘过去积累的知识并确保任务数量增加时的可扩展性。此类问题在自主机器人和视频游戏模拟中普遍存在,尤其是在容易发生拓扑或运动学变化的导航任务中。为了解决这些问题,我们引入了HiSPO,一种专门为从离线数据中进行持续学习的导航设置设计的新型分层框架。我们的方法利用神经网络中不同的策略子空间,以实现灵活且高效的适应新任务,同时保留现有知识。我们通过仔细的实验研究展示了该方法在经典MuJoCo迷宫环境和复杂的视频游戏类导航模拟中的有效性,展示了在经典持续学习指标方面的竞争性表现,特别是在内存使用和效率方面。

英文摘要:

We consider a Continual Reinforcement Learning setup, where a learning agent must continuously adapt to new tasks while retaining previously acquired skill sets, with a focus on the challenge of avoiding forgetting past gathered knowledge and ensuring scalability with the growing number of tasks. Such issues prevail in autonomous robotics and video game simulations, notably for navigation tasks prone to topological or kinematic changes. To address these issues, we introduce HiSPO, a novel hierarchical framework designed specifically for continual learning in navigation settings from offline data. Our method leverages distinct policy subspaces of neural networks to enable flexible and efficient adaptation to new tasks while preserving existing knowledge. We demonstrate, through a careful experimental study, the effectiveness of our method in both classical MuJoCo maze environments and complex video game-like navigation simulations, showcasing competitive performances and satisfying adaptability with respect to classical continual learning metrics, in particular regarding the memory usage and efficiency.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑