In this work, we focus on an analogical reasoning task that contains rich compositional structures, Raven's Progressive Matrices (RPM). To discover compositional structures of the data, we propose the Scattering Compositional Learner (SCL), an architecture that composes neural networks in a sequence. Our SCL achieves state-of-the-art performance on two RPM datasets, with a 48.7% relative improvement on Balanced-RAVEN and 26.4% on PGM over the previous state-of-the-art. We additionally show that our model discovers compositional representations of objects' attributes (e.g., shape color, size), and their relationships (e.g., progression, union). We also find that the compositional representation makes the SCL significantly more robust to test-time domain shifts and greatly improves zero-shot generalization to previously unseen analogies.

中文翻译：

---

散布组合学习者：在类比推理中发现对象、属性、关系

在这项工作中，我们专注于包含丰富组合结构的类比推理任务，Raven 的渐进矩阵 (RPM)。为了发现数据的组成结构，我们提出了 Scattering Compositional Learner (SCL)，这是一种按序列组成神经网络的架构。我们的 SCL 在两个 RPM 数据集上实现了最先进的性能，与之前的最先进技术相比，Balanced-RAVEN 的相对改进为 48.7%，PGM 的相对改进为 26.4%。我们还表明，我们的模型发现了对象属性（例如，形状颜色、大小）及其关系（例如，进展、联合）的组合表示。我们还发现，组合表示使 SCL 对测试时域偏移更加鲁棒，并极大地改善了对以前未见过的类比的零样本泛化。