A Neuro-Symbolic Explainable Machine Learning Model Using Knowledge Graphs

尽管机器学习(ML)方法推动了人工智能(AI)系统的流行和最新进展，但大多数现有模型在解释其推理过程方面存在不足，即无法为某个AI任务背后的推理提供类似人类的理由。这种“可解释”的功能是人工智能系统在真实世界场景中采用和采用的基本要求，因为用户需要信任和理解系统所做的近似和推理。人工智能在具有高社会和经济影响的环境中的应用(如在医疗保健和法律环境中)将需要黑盒人工智能模型朝着系统的方向发展，该系统可以证明、解释潜在的推理过程并与最终用户对话，提供透明的人类可解释的输出。其目标是支持构建复杂的人工智能系统，用于处理问题回答和文本蕴涵等任务，除了预期的输出外，还可以输出有意义的类似人类的解释。研究问题本项目针对以下研究问题：1.知识图(定义的、基于事实的、语篇级别的)能否与微分(神经)归纳逻辑编程(神经符号方法)结合使用以支持可解释的机器学习？2.哪组量化措施可用于评估可解释的机器学习系统？新的工程内容该项目将首次阐明表示大规模背景知识的知识图与微分归纳逻辑编程之间的联系。EPSRC研究领域自然语言处理、机器学习

MotivationDespite the popularity and recent advances in Artificial Intelligence (AI) systems boosted by machine learning (ML) methods, most of the existing models fall short on their ability to explain their reasoning process, i.e., in providing human-like justifications for the reasoning behind a certain AI task. This functionality of "being interpretable" is a fundamental requirement for the adoption and uptake of AI systems in real-world scenarios, as users need to trust and understand the approximations and inferences done by the system. The application of AI in contexts with high social and economic impact (as in health care and legal settings) will require the evolution of black-box AI models in the direction of systems which can justify, explain and dialogue with their end-users about the underlying reasoning process, providing transparent human-interpretable outputs.ApproachThis project aims at developing a neuro-symbolic explainable Machine Learning model using Knowledge graphs. The goal is to support the construction of complex AI systems for addressing tasks such as Question Answering and Text Entailment, which can output meaningful human-like explanations in addition to the expected output. Research QuestionsThe project targets the following research questions:1. Can knowledge graphs (definitional, fact-based, discourse-level) be used in conjunction to differential(neuro) inductive logic programming (the neuro-symbolic approach) to support explainable machine learning?2. Which set of quantitative measures can be used to evaluate explainable machine learning systems? Novel Engineering ContentThe project will articulate for the first-time the connection between knowledge graphs, which represents large-scale background knowledge and differential inductive logic programming.EPSRC Research AreasNatural Language Processing, Machine Learning