使用一致性检验评估超人类模型
Evaluating Superhuman Models with Consistency Checks
AI总结:
本文提出通过一致性检验评估超人类模型的框架,在评估国际象棋、预测事件和法律判决任务中,即使模型表现超人类,也能通过逻辑规则发现其决策不一致性。
摘要:
如果机器学习模型在各种推理或决策任务中达到超人类能力,鉴于人类必然无法作为真实标签的良好代理,我们该如何评估此类模型?本文提出一个通过一致性检验评估超人类模型的框架。我们的前提是,虽然超人类决策的正确性可能无法评估,但如果模型的决策未能满足某些逻辑的、人类可解释的规则,我们仍能暴露其错误。我们在三项任务上实例化了该框架,这些任务由于超人类模型能力或缺失真实标签而难以评估决策正确性:评估国际象棋棋局、预测未来事件以及做出法律判决。我们表明,无论模型在这些任务上的表现如何(可能是超人类的),我们都能发现决策中的逻辑不一致性。例如:国际象棋引擎对语义相同的棋盘给出相反的估值;GPT-4预测体育记录随时间非单调演变;或者AI法官仅在向被告犯罪记录中增加一项重罪后才准予保释。
英文摘要:
If machine learning models were to achieve superhuman abilities at various reasoning or decision-making tasks, how would we go about evaluating such models, given that humans would necessarily be poor proxies for ground truth? In this paper, we propose a framework for evaluating superhuman models via consistency checks. Our premise is that while the correctness of superhuman decisions may be impossible to evaluate, we can still surface mistakes if the model's decisions fail to satisfy certain logical, human-interpretable rules. We instantiate our framework on three tasks where correctness of decisions is hard to evaluate due to either superhuman model abilities, or to otherwise missing ground truth: evaluating chess positions, forecasting future events, and making legal judgments. We show that regardless of a model's (possibly superhuman) performance on these tasks, we can discover logical inconsistencies in decision making. For example: a chess engine assigning opposing valuations to semantically identical boards; GPT-4 forecasting that sports records will evolve non-monotonically over time; or an AI judge assigning bail to a defendant only after we add a felony to their criminal record.
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。