随机空隙：一种马尔可夫框架，用于代理人工智能的预部署可靠性和监督成本审计
The Stochastic Gap: A Markovian Framework for Pre-Deployment Reliability and Oversight-Cost Auditing in Agentic Artificial Intelligence
AI总结：
本文提出一种马尔可夫框架，用于评估代理人工智能在部署前的可靠性及监督成本审计，通过分析业务流程日志，展示了如何通过扩展状态空间来提高决策的统计支持度和经济可控性。
摘要：
组织中的代理人工智能（AI）是一个受可靠性和监督成本约束的连续决策问题。当确定性工作流被动作和工具调用的随机策略取代时，关键问题不在于下一步是否合理，而在于由此产生的轨迹是否在统计上得到支持，局部上无歧义且经济上可控。我们开发了一种测度论马尔可夫框架来处理这一设置。核心量包括状态盲区质量B_n(tau)、状态-动作盲区质量B^SA_{pi,n}(tau)、基于熵的人工在环升级门控，以及工作流访问测度上的预期监督成本身份。我们将在2019年业务流程智能挑战赛的采购-支付日志（251,734个案例，1,595,923个事件，42种不同的工作流动作）上实例化该框架，并从相同过程的按时间顺序80/20分割中构建一个日志驱动的模拟代理。主要经验发现是，一个大型工作流在状态层面可能得到良好的支持，但保留了大量盲区质量在下一步决策中：将操作状态扩展为包含案例上下文、经济规模和行动者类别，将状态空间从42扩展到668，并将状态-动作盲区质量从tau=50时的0.0165提高到tau=1000时的0.1253。在测试分割上，m(s) = max_a pi-hat(a|s)在平均上跟踪实际自主步骤准确性，误差在3.4个百分点以内。相同限定统计可信自主性的量也决定了预期监督负担。该框架在大规模企业采购工作流上得到演示，并设计用于直接应用于有操作事件日志的工程流程。
英文摘要：
Agentic artificial intelligence (AI) in organizations is a sequential decision problem constrained by reliability and oversight cost. When deterministic workflows are replaced by stochastic policies over actions and tool calls, the key question is not whether a next step appears plausible, but whether the resulting trajectory remains statistically supported, locally unambiguous, and economically governable. We develop a measure-theoretic Markov framework for this setting. The core quantities are state blind-spot mass B_n(tau), state-action blind mass B^SA_{pi,n}(tau), an entropy-based human-in-the-loop escalation gate, and an expected oversight-cost identity over the workflow visitation measure. We instantiate the framework on the Business Process Intelligence Challenge 2019 purchase-to-pay log (251,734 cases, 1,595,923 events, 42 distinct workflow actions) and construct a log-driven simulated agent from a chronological 80/20 split of the same process. The main empirical finding is that a large workflow can appear well supported at the state level while retaining substantial blind mass over next-step decisions: refining the operational state to include case context, economic magnitude, and actor class expands the state space from 42 to 668 and raises state-action blind mass from 0.0165 at tau=50 to 0.1253 at tau=1000. On the held-out split, m(s) = max_a pi-hat(a|s) tracks realized autonomous step accuracy within 3.4 percentage points on average. The same quantities that delimit statistically credible autonomy also determine expected oversight burden. The framework is demonstrated on a large-scale enterprise procurement workflow and is designed for direct application to engineering processes for which operational event logs are available.
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。
