arXiv · cs.AI · Vincess Dongmo, Astride Melvin Fokam Ninyim · 10 小时前模型评分37
多关系图学习研究：何时该信任一条关系？操纵成本、暴露度与信任陷阱
When Is a Relation Worth Trusting? Manipulation Costs, Exposure, and the Trust Trap in Multi-Relational Graph Learning
导读
一项多关系图学习研究提出按度数门控（degree gating）的关系使用方式，证明当部分节点在某一关系上被孤立、且攻击者无力打开门控时，它严格优于对该关系的任何选择或重加权。
分类置信度
中 · 待人工复核
- 只有单一媒体来源，需要人工复核。
来源级别：媒体与个人
事件类型：研究发布
证据摘录
- “View PDF Abstract: Multi-relational graph detectors learn from clean data how much to trust each relation, but an adversary pays a different price to manipulate each relation and each action; taking such costs as given, we ask when using a relation is worse than ignoring it.”
- “Without cost information, the safe choice is to abstain.”
- “with at most four relations it mostly abstains, as validation alone would, and its cost-aware proposals help when relations are many, so far shown on synthetic graphs.”
来源：arXiv · cs.AI · arxiv.org