arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2606.04779cs.AImath.CO
收藏

基于树的人机交互中多智能体互补性形式化

Tree-Based Formalization of Multi-Agent Complementarity in Human-AI Interactions

Institute of Biomedical Ethics and History of Medicine, University of Zurich(伦理与医学史研究所,苏黎世大学);SUPSI, Dalle Molle Institute for Artificial Intelligence (IDSIA)(SUPSI,达勒莫利人工智能研究所);ETH Zurich(苏黎世联邦理工学院)

Andrea Ferrario

AI总结:

本文提出一种基于树的形式化框架,通过有序智能体角色配置和平面二叉树表示人机交互协议,证明互补性在回归中可实现,但在分类中受限于局部聚合和损失函数的自然条件。

摘要:

互补性是指人机交互(HAI)的表现优于其成员中最佳预测基准的情况。尽管这一概念在HAI研究中至关重要,但关于互补性的形式化工作仍然有限。现有框架未能建模智能体的预测如何组合成对工作流敏感的多智能体协议。我们通过引入基于树的多智能体HAI互补性形式化来填补这一空白。一个HAI协议由一个有序的智能体角色配置以及一棵有根平面二叉树表示,树的叶子由预测向量装饰。沿树递归评估一个局部二元组合规则,产生相对于逐点最小预言基准的树相对互补性泛函。我们证明了四个结果。第一,基于选择器的HAI(包括自我或AI依赖)无法实现互补性,无论任务、损失或预测质量如何。第二,在平方损失下的回归中,互补性等价于与真实向量之间的欧几里得距离最小化;对于$N=2$,最优线性池化权重具有封闭形式并具有残差校正解释。第三,在线性局部组合下,每个协议树定义了叶子权重单纯形上的重心坐标图;协议树的Tamari覆盖重新参数化保持互补性,对于$N=4$,它们满足五边形恒等式。第四,在二元分类中,在端点单调损失(包括标准Bregman和许多有限伯努利$f$散度损失)下,没有内部局部组合能实现互补性;在交叉熵下的多类聚合中存在类似障碍。总之,我们的框架表明,互补性在多智能体回归中是可实现的,但在分类中,在局部聚合和损失函数的自然条件下受到阻碍。

英文摘要:

Predictive tasks constitute subprocedures within human--AI interactions (HAIs) involving professionals and AI systems. Complementarity occurs when an HAI outperforms a benchmark defined by its constituent agents. Existing formalizations, however, focus on a single human and a single AI, leaving multi-agent interactions insufficiently characterized. We introduce a tree-based mathematical framework for multi-agent complementarity. Ordered agent-role configurations decorate the leaves of rooted planar binary trees and local aggregation rules generate protocol outputs evaluated against either a best-average or a pointwise-min benchmark. We establish four results. First, under linear pooling, every protocol tree provides a bijective coordinate system for the interior of the simplex of global leaf weights. Arbitrary trees with the same ordered leaves admit output- and complementarity-preserving reparameterizations, realized locally through path-independent Tamari rotations. However, HAI constraints can violate admissibility preservation between equivalent protocols. Second, for regression under squared loss, we characterize complementarity through residual correction and derive optimal linear-pooling weights. We also characterize protocol-indifference loci under identical local parameters. Third, selector-based protocols cannot achieve complementarity relative to the pointwise-min benchmark, independently of task and loss. Fourth, in binary classification, HAIs cannot achieve complementarity under endpoint-monotone losses and the pointwise-min benchmark, with an analogous obstruction for multiclass cross-entropy. Numerical illustrations examine the main claims. We provide a mathematical foundation for analyzing multi-agent predictive subprocedures within HAIs, establishing a first step toward understanding how interaction, aggregation, and real-world constraints affect HAI performance.

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。

↑