论文标题
在训练集之外概括：神经网络何时可以学习身份效果？
Generalizing Outside the Training Set: When Can Neural Networks Learn Identity Effects?
论文作者
论文摘要
通常在语言和其他认知领域中，无论一个对象的两个组成部分是否相同确定是否形成很好。我们称此类约束身份效应。在开发一个从示例中学习良好的系统时，在识别效果中构建非常容易。但是，如果没有明确的指导，可以从数据中学到身份效应吗？我们提供了一个简单的框架，可以严格证明满足简单标准的算法无法做出正确的推论。然后，我们表明，一类广泛的算法，包括带有标准体系结构的深神经网络和带有反向传播的培训的培训，可以满足我们的标准，取决于输入的编码。最后，我们通过计算实验来证明我们的理论，在这些实验中，我们探讨了不同输入编码对算法推广到新输入能力的影响。
Often in language and other areas of cognition, whether two components of an object are identical or not determine whether it is well formed. We call such constraints identity effects. When developing a system to learn well-formedness from examples, it is easy enough to build in an identify effect. But can identity effects be learned from the data without explicit guidance? We provide a simple framework in which we can rigorously prove that algorithms satisfying simple criteria cannot make the correct inference. We then show that a broad class of algorithms including deep neural networks with standard architecture and training with backpropagation satisfy our criteria, dependent on the encoding of inputs. Finally, we demonstrate our theory with computational experiments in which we explore the effect of different input encodings on the ability of algorithms to generalize to novel inputs.
