arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2403.08222cs.LGcs.AI
收藏

存在对抗性专家的鲁棒决策聚合

Robust Decision Aggregation with Adversarial Experts

School of Computer Science, Peking University(北京大学计算机学院)

Yongkang Guo, Yuqing Kong

AI总结:

本文研究存在对抗性专家时的鲁棒决策聚合问题,提出截断均值聚合器及其随机版本,证明其在L1和L2损失下最优,并扩展到多状态设定,同时给出更一般设定下的负结果。

摘要:

我们考虑在同时存在诚实专家和对抗性专家情况下的鲁棒聚合问题。诚实专家会如实报告其私有信号,而对抗性专家可以任意报告。我们假设专家在边际意义上是对称的,即他们共享相同的公共先验和边际后验。规则制定者需要设计一个聚合器,在不知道底层信息结构或对抗策略的情况下,根据这些专家的报告预测真实世界状态。我们的目标是找到最优聚合器,使其在最坏信息结构和对抗策略下输出最小化遗憾的预测。遗憾定义为聚合器与一个基准之间的期望损失之差,该基准在给定信息结构和诚实专家报告的情况下进行最优聚合。我们聚焦于二值状态和二值报告。在 L1 损失下,我们证明截断均值聚合器是最优的。当最多存在 k 个对抗者时,该聚合器丢弃 k 个最低和 k 个最高的报告值,并对剩余值取平均。对于 L2 损失,最优聚合器是分段线性函数。当对抗者比例的上界由专家先验和后验决定的一个值所限制时,所有这些最优性均成立。遗憾仅取决于对抗者比例,而不取决于其总数。对于输出决策的硬聚合器,我们证明截断均值的随机版本在 L1 和 L2 下都是最优的。该聚合器在丢弃 k 个最低和 k 个最高的报告值后,随机遵循一个剩余值。我们将硬聚合器扩展到多状态设定。我们在集成学习任务中对所提出的聚合器进行了数值评估。我们还得到了在更广泛信息结构和报告空间下一般对抗聚合问题的负结果。

英文摘要:

We consider a robust aggregation problem in the presence of both truthful and adversarial experts. The truthful experts will report their private signals truthfully, while the adversarial experts can report arbitrarily. We assume experts are marginally symmetric in the sense that they share the same common prior and marginal posteriors. The rule maker needs to design an aggregator to predict the true world state from these experts' reports, without knowledge of the underlying information structures or adversarial strategies. We aim to find the optimal aggregator that outputs a forecast minimizing regret under the worst information structure and adversarial strategies. The regret is defined by the difference in expected loss between the aggregator and a benchmark who aggregates optimally given the information structure and reports of truthful experts. We focus on binary states and reports. Under L1 loss, we show that the truncated mean aggregator is optimal. When there are at most k adversaries, this aggregator discards the k lowest and highest reported values and averages the remaining ones. For L2 loss, the optimal aggregators are piecewise linear functions. All the optimalities hold when the ratio of adversaries is bounded above by a value determined by the experts' priors and posteriors. The regret only depends on the ratio of adversaries, not on their total number. For hard aggregators that output a decision, we prove that a random version of the truncated mean is optimal for both L1 and L2. This aggregator randomly follows a remaining value after discarding the $k$ lowest and highest reported values. We extend the hard aggregator to multi-state setting. We evaluate our aggregators numerically in an ensemble learning task. We also obtain negative results for general adversarial aggregation problems under broader information structures and report spaces.

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。

↑