Conditional Neural Processes 用于动作模仿中自监督目标发现的研究
Conditional Neural Processes for Self-Supervised Goal Discovery in Action Imitation
AI 导读
受婴儿通过模仿学习新任务的能力启发，研究评估了 Conditional Neural Processes 能否自主学习蕴含动作底层目标的行为表征。结果表明，在隐空间中加入注意力机制可在动作序列观测早期催生稳健的目标表征，使其有望成为机器人自监督通用学习器。该工作已在 IROS2023 认知机器人世界模型与预测编码研讨会海报环节展示。
正文 · 原文
Abstract:Inspired by infants' ability to learn novel tasks by imitating others, in this work we evaluate to what extent Conditional Neural Processes can autonomously learn representations of actions that are informative about their underlying goal. We show that the addition of an attention mechanism in the hidden space fosters the emergence of robust goal representations early on in the observation of an action sequence, making it a promising model for a robotic self-supervised general learner.
| Comments: | Presented at the poster session of the Workshop on World Models and Predictive Coding in Cognitive Robotics @ IROS2023 (this https URL) |
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.32688 [cs.RO] |
| (or arXiv:2610.32688v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2610.32688 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Marco Gabriele Fedozzi [view email]
[v1]
Fri, 23 Oct 2026 14:19:34 UTC (1,598 KB)
来源：arXiv cs.AI · arxiv.org