面向交互式指令跟随代理的在线持续学习
Online Continual Learning For Interactive Instruction Following Agents
AI总结：
针对具身智能体在线持续学习，提出行为增量与环境增量两种设置，并设计无需任务边界的置信度感知移动平均方法，显著优于现有方法。
摘要：
在通过语言指令学习执行日常任务的具身智能体研究中，文献普遍假设智能体在初始阶段一次性学习所有训练数据。我们认为这种学习场景不太现实，因为机器人智能体应当在其探索和感知世界的过程中持续地学习世界。为了向更真实的具身智能体学习场景迈进一步，我们为具身智能体提出了两种持续学习设置：学习新行为（行为增量学习，Behavior-IL）和学习新环境（环境增量学习，Environment-IL）。对于这些任务，先前基于“数据先验”的持续学习方法会为过去的任务保留逻辑值。然而，存储的信息往往是学习不充分的信息，并且需要任务边界信息，而任务边界信息并不总是可用的。在此，我们提出在训练过程中无需任务边界信息（即无任务）的情况下，基于置信度分数以移动平均的方式更新这些逻辑值，称为置信度感知移动平均（CAMA）。在所提出的Behavior-IL和Environment-IL设置中，我们简单的CAMA在经验验证中显著优于先前的最先进方法。包含代码的项目页面为https://github.com/snumprlab/cl-alfred。
英文摘要：
In learning an embodied agent executing daily tasks via language directives, the literature largely assumes that the agent learns all training data at the beginning. We argue that such a learning scenario is less realistic since a robotic agent is supposed to learn the world continuously as it explores and perceives it. To take a step towards a more realistic embodied agent learning scenario, we propose two continual learning setups for embodied agents; learning new behaviors (Behavior Incremental Learning, Behavior-IL) and new environments (Environment Incremental Learning, Environment-IL) For the tasks, previous 'data prior' based continual learning methods maintain logits for the past tasks. However, the stored information is often insufficiently learned information and requires task boundary information, which might not always be available. Here, we propose to update them based on confidence scores without task boundary information during training (i.e., task-free) in a moving average fashion, named Confidence-Aware Moving Average (CAMA). In the proposed Behavior-IL and Environment-IL setups, our simple CAMA outperforms prior state of the art in our empirical validations by noticeable margins. The project page including codes is https://github.com/snumprlab/cl-alfred.
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。
