论文标题
通过推理潜在实例来验证树的合奏
Verifying Tree Ensembles by Reasoning about Potential Instances
论文作者
论文摘要
想象一下，能够向黑匣子模型提出问题，例如“存在哪种对抗示例？”，“特定属性会对模型的预测产生不成比例的影响？或“对于部分描述的例子，可以做出什么样的预测？”如果您的部分描述与数据中观察到的任何示例不符，那么最后一个问题尤其重要，因为它提供了有关模型如何推断出不见数据的洞察力。这些功能将非常有帮助，因为它们将使用户更好地理解模型的行为，尤其是与稳健性，公平性和偏见等问题有关。在本文中，我们为树木合奏提出了一种方法。通常，由于此任务是棘手的，因此我们提出了一种策略，即（1）考虑到要简化问题的问题，该策略可以修剪一部分输入空间； （2）遵循一种划分和征服方法，该方法是增量的，并且可以始终返回一些答案，并指示输入域的哪些部分仍然不确定。我们方法的有用性显示在各种用例中。
Imagine being able to ask questions to a black box model such as "Which adversarial examples exist?", "Does a specific attribute have a disproportionate effect on the model's prediction?" or "What kind of predictions could possibly be made for a partially described example?" This last question is particularly important if your partial description does not correspond to any observed example in your data, as it provides insight into how the model will extrapolate to unseen data. These capabilities would be extremely helpful as they would allow a user to better understand the model's behavior, particularly as it relates to issues such as robustness, fairness, and bias. In this paper, we propose such an approach for an ensemble of trees. Since, in general, this task is intractable we present a strategy that (1) can prune part of the input space given the question asked to simplify the problem; and (2) follows a divide and conquer approach that is incremental and can always return some answers and indicates which parts of the input domains are still uncertain. The usefulness of our approach is shown on a diverse set of use cases.
