Attention on Abstract Visual Reasoning
Lukas Hahne Timo Lüddecke Florentin Wörgötter David Kappel Correspond to: Bernstein Center for Computational Neuroscience III Physikalisches Institut-Biophysik, Georg-August Universität Göttingen, Germany
Abstract
Attention mechanisms have been boosting the performance of deep learning models on a wide range of applications, ranging from speech understanding to program induction. However, despite experiments from psychology which suggest that attention plays an essential role in visual reasoning, the full potential of attention mechanisms has so far not been explored to solve abstract cognitive tasks on image data. In this work, we propose a hybrid network architecture, grounded on self-attention and relational reasoning. We call this new model Attention Relation Network (ARNe). ARNe combines features from the recently introduced Transformer and the Wild Relation Network (WReN). We test ARNe on the Procedurally Generated Matrices (PGMs) datasets for abstract visual reasoning. ARNe excels the WReN model on this task by 11.28 ppt. Relational concepts between objects are efficiently learned demanding only $35\%$ of the training samples to surpass reported accuracy of the base line model. Our proposed hybrid model, represents an alternative on learning abstract relations using self-attention and demonstrates that the Transformer network is also well suited for abstract visual reasoning.
中文速览
抽象视觉推理考验模型理解隐含规则的能力，现有方法（如WReN）虽擅长关系推理却缺乏注意力机制，而Transformer擅长自注意力但从未被用于此类任务。作者将两者结合，提出了"注意力关系网络"（Attention Relation Network，ARNe），用Transformer的自注意力模块替代WReN中的关系聚合部分，并在程序生成矩阵（Procedurally Generated Matrices，PGM）数据集上进行训练和评测。ARNe以88.18%的准确率完成抽象视觉推理任务，比此前最优模型高出11.28个百分点，且只需全量训练数据的35%就能超越基线模型的最终水平。这项工作证明了自注意力机制不仅适用于语言理解，同样能有效提升机器在抽象视觉推理这类考验流体智力任务上的表现，并大幅降低对标注样本的需求。
原文 arXiv:1911.05990;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1911.05990v1