A Framework for End-to-End Learning on Semantic Tree-Structured Data

Abstract

While learning models are typically studied for inputs in the form of a fixed dimensional feature vector, real world data is rarely found in this form. In order to meet the basic requirement of traditional learning models, structural data generally have to be converted into fix-length vectors in a handcrafted manner, which is tedious and may even incur information loss. A common form of structured data is what we term "semantic tree-structures", corresponding to data where rich semantic information is encoded in a compositional manner, such as those expressed in JavaScript Object Notation (JSON) and eXtensible Markup Language (XML). For tree-structured data, several learning models have been studied to allow for working directly on raw tree-structure data, however such learning models are limited to either a specific tree-topology or a specific tree-structured data format, e.g., synthetic parse trees. In this paper, we propose a novel framework for end-to-end learning on generic semantic tree-structured data of arbitrary topologies and heterogeneous data types, such as data expressed in JSON, XML and so on. Motivated by the works in recursive and recurrent neural networks, we develop exemplar neural implementations of our framework for the JSON format. We evaluate our approach on several UCI benchmark datasets, including ablation and data-efficiency studies, and on a toy reinforcement learning task. Experimental results suggest that our framework yields comparable performance to use of standard models with dedicated feature-vectors in general, and even exceeds baseline performance in cases where compositional nature of the data is particularly important. The source code for a JSON-based implementation of our framework along with experiments can be downloaded at https://github.com/EndingCredits/json2vec.

Deep Reinforcement Learning for IoT Interoperability

“…During the last stages of our research, we came to know about the work of Woof and Chen who provide a framework for end-to-end learning on treestructured data [19]. Their work possesses some parallels with ours, however differs on critical aspects.…”

Structured Context and High-Coverage Grammar for Conversational Question Answering over Knowledge Graphs

“…Tai et al (2015) introduced Tree-LSTM for computing tree embeddings bottom-up. It has then been applied for many tasks, including computer program translation , semantic tree structure learning (such as JSON or XML) (Woof and Chen, 2020) and supervised KG-QA tasks (Tong et al, 2019;Zafar et al, 2019;Athreya et al, 2021). In the latter context, Tree-LSTM is used to model the syntactic structure of the question.…”

Structured Context and High-Coverage Grammar for Conversational Question Answering over Knowledge Graphs

“…Tai et al (2015) introduced Tree-LSTM for computing tree embeddings bottom-up. It has then been applied for many tasks, including computer program translation , se-mantic tree structure learning (such as JSON or XML) (Woof and Chen, 2020) and supervised KG-QA tasks (Tong et al, 2019;Zafar et al, 2019;Athreya et al, 2020). In the latter context, Tree-LSTM is used to model the syntactic structure of the question.…”