CausalLoss-Fin：将金融智能体的损失归因于决策与基础设施故障
CausalLoss-Fin: Attributing Financial-Agent Loss to Decisions and Infrastructure Faults
浏览论文内容
中文总结 AI 辅助
本文提出CausalLoss-Fin，通过同时干预智能体决策与基础设施故障，利用伸缩恒等式和Shapley值将金融智能体损失精确分解，证明仅归因于决策会错失全部可恢复损失。
中文摘要 AI 辅助
当处理支付异常的智能体发生资金损失时，本文所比较的智能体步骤归因方法会将其归因于该智能体的某个动作。即使结算消息被丢弃、智能体从未有机会处理时，这些方法仍会如此归因：它们对智能体动作进行干预，但未将基础设施故障暴露为可干预变量，因此它们解释的每一美元损失都被归咎于某个决策。我们采用一个故障过程明确且可重放的基准，将每个回合的实际投递时间表分解为命名的、可单独修复的消息，并对智能体的选择和基础设施的选择同时进行干预。一个伸缩恒等式将任意策略的损失精确地分解为三部分：基础设施效应、相对于最佳可实施策略的策略差异，以及参考策略残差。其中两部分可能为负，因此它们都不是份额；随后Shapley值将第一部分划分为对单个消息的有符号分配。一个结果是结构性的，无需语料库即可成立：仅智能体基线无法识别任何基础设施原因，因为其模型不包含能命名该原因的变量。3个策略下植入的545个回合所衡量的正是该后果的规模。它错误地将100%的基础设施回合归类，并将114,383.40美元记入智能体。修复其所命名的原因可恢复可用损失的0.0%；修复最小充分集可恢复100.0%。逐个对消息进行评分不仅不精确：27.8%（95%置信区间：23.3%–32.3%）的回合无法加性分解。我们评估的是确定性程序化策略而非语言模型智能体，这使重放精确，但也将外部效度限制于随机智能体。这些发生率数值是该生成器的属性，而非现场比率。
英文摘要
When an agent handling a payment exception loses money, the agent-step attribution methods this paper compares against will name one of its actions. They will do so even when a settlement message was dropped and the agent never had a chance: they intervene on agent actions and do not expose infrastructure faults as intervenable variables, so every dollar they explain is charged to a decision. We take a benchmark whose fault process is explicit and replayable, decompose each episode's realised delivery schedule into named, individually repairable messages, and intervene on both the agent's choices and the infrastructure's. A telescoping identity splits any policy's loss exactly three ways: an infrastructure effect, a policy differential against the best implementable policy, and a reference-policy residual. Two of the three can be negative, so none is a share; Shapley then divides the first into signed allocations over individual messages. One result is structural and needs no corpus: an agent-only baseline identifies no infrastructure cause, because its model contains no variable that could name one. What 545 planted episodes across 3 policies measure is the size of that consequence. It misfiles 100% of infrastructure episodes and charges $114,383.40 to the agent. Repairing what it names recovers 0.0% of the available loss; repairing a minimal sufficient set recovers 100.0%. Scoring messages one at a time is not merely imprecise: 27.8% (95% CI: 23.3--32.3%) of episodes do not decompose additively. We evaluate deterministic programmatic policies rather than language-model agents, which is what makes replay exact and which limits external validity to stochastic agents. The prevalence figures are properties of this generator, not field rates.
