Hagenbuchner M, Sperduti A, Tsoi Ah Chung
Fac. of Informatics, Univ. of Wollongong, NSW, Australia.
IEEE Trans Neural Netw. 2003;14(3):491-505. doi: 10.1109/TNN.2003.810735.
Recent developments in the area of neural networks produced models capable of dealing with structured data. Here, we propose the first fully unsupervised model, namely an extension of traditional self-organizing maps (SOMs), for the processing of labeled directed acyclic graphs (DAGs). The extension is obtained by using the unfolding procedure adopted in recurrent and recursive neural networks, with the replicated neurons in the unfolded network comprising of a full SOM. This approach enables the discovery of similarities among objects including vectors consisting of numerical data. The capabilities of the model are analyzed in detail by utilizing a relatively large data set taken from an artificial benchmark problem involving visual patterns encoded as labeled DAGs. The experimental results demonstrate clearly that the proposed model is capable of exploiting both information conveyed in the labels attached to each node of the input DAGs and information encoded in the DAG topology.
神经网络领域的最新进展产生了能够处理结构化数据的模型。在此，我们提出了首个完全无监督的模型，即传统自组织映射（SOM）的扩展，用于处理带标签的有向无环图（DAG）。这种扩展是通过使用循环和递归神经网络中采用的展开过程实现的，展开网络中的复制神经元构成一个完整的SOM。这种方法能够发现包括由数值数据组成的向量在内的对象之间的相似性。通过利用一个相对较大的数据集，该数据集取自一个涉及编码为带标签DAG的视觉模式的人工基准问题，对模型的能力进行了详细分析。实验结果清楚地表明，所提出的模型能够利用输入DAG每个节点上附加标签中传达的信息以及DAG拓扑结构中编码的信息。
