Abductive Knowledge Induction From Raw Data
Abductive Knowledge Induction From Raw Data
复制标题
DOI：
10.24963/ijcai.2021/254
复制
发表时间：
2020-10
期刊：
影响因子：
--
通讯作者：
Wang-Zhou Dai;S. Muggleton
中科院分区：
文献类型：
--
作者：
Wang-Zhou Dai;S. Muggleton
对于许多具有原始输入的推理繁重任务，设计适当的端到端管道来制定问题解决过程具有挑战性。一些现代人工智能系统，例如神经符号学习，将管道分为子符号感知和符号推理，试图同时利用数据驱动的机器学习和知识驱动的问题解决。然而，这些系统由于两个组件之间的接口而导致计算复杂度呈指数级增长，其中子符号学习模型缺乏直接监督，而符号模型缺乏准确的输入事实。因此，他们通常专注于学习具有完整符号知识库的子符号模型，同时避免了一个关键问题：知识从哪里来？在本文中，我们提出了溯因元解释学习（MetaAbd），它将溯因和归纳结合起来，从原始数据中共同学习神经网络和逻辑理论。实验结果表明，MetaAbd 不仅在预测准确性和数据效率方面优于对比系统，而且还诱导出可在后续学习任务中重新用作背景知识的逻辑程序。据我们所知，MetaAbd 是第一个可以从头开始联合学习神经网络并通过谓词发明导出递归一阶逻辑理论的系统。
For many reasoning-heavy tasks with raw inputs, it is challenging to design an appropriate end-to-end pipeline to formulate the problem-solving process. Some modern AI systems, e.g., Neuro-Symbolic Learning, divide the pipeline into sub-symbolic perception and symbolic reasoning, trying to utilise data-driven machine learning and knowledge-driven problem-solving simultaneously. However, these systems suffer from the exponential computational complexity caused by the interface between the two components, where the sub-symbolic learning model lacks direct supervision, and the symbolic model lacks accurate input facts. Hence, they usually focus on learning the sub-symbolic model with a complete symbolic knowledge base while avoiding a crucial problem: where does the knowledge come from? In this paper, we present Abductive Meta-Interpretive Learning (MetaAbd) that unites abduction and induction to learn neural networks and logic theories jointly from raw data. Experimental results demonstrate that MetaAbd not only outperforms the compared systems in predictive accuracy and data efficiency but also induces logic programs that can be re-used as background knowledge in subsequent learning tasks. To the best of our knowledge, MetaAbd is the first system that can jointly learn neural networks from scratch and induce recursive first-order logic theories with predicate invention.