学习难以改变的解释
Learning explanations that are hard to vary
摘要：
在本文中，我们研究了在深度学习的背景下“好的解释很难改变”的原则。我们表明，平均梯度的例子-类似于一个逻辑或模式-可以有利于记忆和“拼凑”的解决方案，缝合在一起不同的策略，而不是确定不变。为了检验这一点，我们首先形式化的损失面的最小值的一致性的概念，它的措施，在何种程度上最小值出现时，只有汇集的例子。然后，我们提出并实验验证了一个简单的替代算法的基础上的逻辑与，专注于不变性和防止记忆在一组现实世界的任务。最后，使用一个明确区分不变和虚假机制的合成数据集，我们剖析了学习信号，并将这种方法与成熟的正则化器进行了比较。
英文摘要：
In this paper, we investigate the principle that `good explanations are hard to vary' in the context of deep learning. We show that averaging gradients across examples -- akin to a logical OR of patterns -- can favor memorization and `patchwork' solutions that sew together different strategies, instead of identifying invariances. To inspect this, we first formalize a notion of consistency for minima of the loss surface, which measures to what extent a minimum appears only when examples are pooled. We then propose and experimentally validate a simple alternative algorithm based on a logical AND, that focuses on invariances and prevents memorization in a set of real-world tasks. Finally, using a synthetic dataset with a clear distinction between invariant and spurious mechanisms, we dissect learning signals and compare this approach to well-established regularizers.
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。
