超越模仿：迈向模仿学习中的终身适应性
Beyond Mimicry: Toward Lifelong Adaptability in Imitation Learning
AI总结：
本文提出通过重新定义模仿学习目标，实现终身适应性，强调通过组合基本行为要素提升在动态环境中的适应能力。
摘要：
模仿学习正处于十字路口：尽管数十年来取得了进步，当前的模仿学习代理仍然是复杂的记忆机器，擅长回放但当上下文变化或目标演变时却失败。本文认为这种失败不是技术问题而是基础问题：模仿学习已被优化为错误的目标。我们提出一个研究议程，将成功从完美回放重新定义为组合适应性。这种适应性取决于学习行为基本要素一次，并通过新的上下文重新组合而不重新训练。我们建立了组合泛化指标，提出了混合架构，并概述了跨学科研究方向，借鉴认知科学和文化进化。将适应性嵌入模仿学习核心的代理因此具备在开放世界中运作的基本能力。
英文摘要：
Imitation learning stands at a crossroads: despite decades of progress, current imitation learning agents remain sophisticated memorisation machines, excelling at replay but failing when contexts shift or goals evolve. This paper argues that this failure is not technical but foundational: imitation learning has been optimised for the wrong objective. We propose a research agenda that redefines success from perfect replay to compositional adaptability. Such adaptability hinges on learning behavioural primitives once and recombining them through novel contexts without retraining. We establish metrics for compositional generalisation, propose hybrid architectures, and outline interdisciplinary research directions drawing on cognitive science and cultural evolution. Agents that embed adaptability at the core of imitation learning thus have an essential capability for operating in an open-ended world.
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。
