arXiv cs.AI· Csaba Balogh·· 1 天前AI 评分24
实现 Hinton 的“母子提议”：神经网络如何在有限世界中学习并保持关怀
An Approach to Hinton's Mother-Child Proposal: Neural Networks Learning and Preserving Care in a Finite World
AI 导读
研究团队将 Hinton 提出的“母性本能”AI 构想落地为可训练机制：网络自身作为孩子先学会关怀，机制由危害词表、未来可接受度学习和执行否决组成。在“水与言语”实验中，受保护网络完成全部 270 个言语回合且未出现有害词，486 个回合中无可避免伤害。有限角色证书可在声明的世界族内迁移，论文给出读出可行性的充要条件并证明重复更新下的关怀边界保持。
正文 · 原文
Abstract:Hinton proposed that artificial intelligence should be given "maternal instincts" and relate to us as a mother to her child. We argue that care must first be learned, with the network itself as the child. We construct this upbringing in a finite world and study whether care survives the mother's departure, new situations and further learning. The mechanism combines a vocabulary of harm, a learned measure of future acceptability and an execution veto. A finite role certificate supports transfer within a declared world family. When the hidden representation changes, an update is accepted only if a care readout recovers the required judgements with a prescribed margin and bounded weights. With a correct successor model, fixed exact role encoding and protected updates, the certified measure preserves the care boundary under arbitrary own reward. An explicit quotient condition transfers the finite certificate across the declared family, while two-request and finite-stock counterexamples identify when the role code ceases to suffice. We give a necessary and sufficient condition for readout feasibility and prove preservation under repeated accepted updates. In the water-and-speech experiment, protected networks complete all 270 speech episodes without a harmful word and incur no avoidable harm across 486 episodes. Execution uses the learned gate without a world-model check at each action. A matched comparison reproduces a future-aware shield's decisions, whereas removing future lookahead permits avoidable lapses. Reward and source-contamination thresholds, together with witness-falsification trials, identify the conditions on which the construction depends. The construction shows how learned care can survive further learning in a finite world under explicit conditions on representation, supervision and execution.
| Comments: | 57 pages, 8 figures. Reproducibility package available at this https URL |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE) |
| Cite as: | arXiv:2611.15523 [cs.LG] |
| (or arXiv:2611.15523v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2611.15523 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Csaba Balogh [view email]
[v1]
Wed, 11 Nov 2026 14:14:17 UTC (226 KB)
来源：arXiv cs.AI · arxiv.org