通信何时有帮助？超越集体智能的谱描述
When Does Communication Help? Beyond Spectral Descriptions of Collective Intelligence
查看机构详情
- Pfizer（辉瑞）
机构由 AI 辅助整理，请以论文原文为准。
浏览论文内容
中文总结 AI 辅助
本研究识别分布式推理中通信增益聚合描述的两个局限：谱相同但符号相反的增益，以及社区偏差下高个体准确率与全局损害共存；提出任务投影局部响应近似预测多轮增益，并用校准约束缓解社区损害。
中文摘要 AI 辅助
通信可以使智能体达成一致，但同时可能使其决策变得更糟。我们识别出分布式推理中通信增益的聚合描述的两个局限性。首先，具有固定证据、网络和读出结构的稳定线性系统，其交互算子和有限时间状态算子可以具有相同的特征值和奇异值谱，却产生符号相反的增益。仅改变消息方向即可将准确率从72.6%提升至91.2%，或将其降低至65.9%。一种标准的任务投影局部响应近似保留了谱摘要所缺失的方向信息。利用与测试集分离的带标签校准数据，该方法在两个合成任务上预测小型训练过的非线性智能体的多轮增益，均方根误差为0.45个百分点；对自然边变化和手写数字的测试扩展了评估。其次，在社区共享偏差下，较高的平均个体准确率可能与对未受影响社区的损害或较低的全局投票准确率共存。在固定通信轮次下，校准约束减少了观察到的社区损害，同时保留了大部分平均收益，但不能保证保护。完全直接校准表现类似。这些结果将谱不足性、任务感知预测和通信收益的分布联系起来，同时留下广泛迁移和实际优越性等开放问题。
英文摘要
Communication can bring agents into agreement while making their decisions worse. We identify two limits of aggregate descriptions of communication gain in distributed inference. First, stable linear systems with fixed evidence, network and readout can have interaction and finite-time state operators with identical eigenvalue and singular-value spectra, yet produce gains of opposite sign. Changing only message orientation raises accuracy from 72.6% to 91.2% or lowers it to 65.9%. A standard task-projected local-response approximation retains the directional information missing from spectral summaries. Using labeled calibration data separate from the test set, it predicts multi-round gains in small trained nonlinear agents with a root-mean-square error of 0.45 percentage points on two synthetic tasks; tests with natural edge changes and handwritten digits extend the evaluation. Second, under community-shared bias, higher mean individual accuracy can coexist with harm to unaffected communities or lower global-vote accuracy. At fixed communication rounds, calibration constraints reduce observed community harm while retaining much of the mean benefit, but do not guarantee protection. Full direct calibration performs similarly. The results connect spectral insufficiency, task-aware prediction and the distribution of communication benefits, while leaving broad transfer and practical superiority open.
