Learning Accurate and Interpretable Decision Rule Sets from Neural Networks
Learning Accurate and Interpretable Decision Rule Sets from Neural Networks
复制标题
DOI：
10.1609/aaai.v35i5.16555
复制
发表时间：
2021-03
期刊：
影响因子：
--
通讯作者：
Litao Qiao;Weijia Wang;Bill Lin
中科院分区：
文献类型：
--
作者：
Litao Qiao;Weijia Wang;Bill Lin
本文提出了一种新的范式，用于学习一组独立的逻辑规则的析取范式作为分类的可解释模型。我们认为学习一个可解释的决策规则集的问题是在一个特定的，但非常简单的两层架构中训练神经网络。第一层中的每个神经元在训练后直接映射到可解释的if-then规则，第二层中的输出神经元直接映射到第一层规则的析取以形成决策规则集。我们在第一个规则层中对神经元的表示使我们能够在决策规则中对特征的正关联和负关联进行编码。可以利用最先进的神经网络训练方法来学习高度准确的分类模型。此外，我们提出了一种基于稀疏性的正则化方法，以平衡分类精度和简单性之间的派生规则。我们的实验结果表明，我们的方法可以生成更准确的决策规则集比其他国家的最先进的规则学习算法具有更好的准确性和简单性的权衡。此外，与随机森林和全精度深度神经网络等不可解释的黑箱机器学习方法相比，我们的方法可以轻松找到具有可比预测性能的可解释决策规则集。
This paper proposes a new paradigm for learning a set of independent logical rules in disjunctive normal form as an interpretable model for classification. We consider the problem of learning an interpretable decision rule set as training a neural network in a specific, yet very simple two-layer architecture. Each neuron in the first layer directly maps to an interpretable if-then rule after training, and the output neuron in the second layer directly maps to a disjunction of the first layer rules to form the decision rule set. Our representation of neurons in this first rules layer enables us to encode both the positive and the negative association of features in a decision rule. State-of-the-art neural net training approaches can be leveraged for learning highly accurate classification models. Moreover, we propose a sparsity-based regularization approach to balance between classification accuracy and the simplicity of the derived rules. Our experimental results show that our method can generate more accurate decision rule sets than other state-of-the-art rule-learning algorithms with better accuracy-simplicity trade-offs. Further, when compared with uninterpretable black-box machine learning approaches such as random forests and full-precision deep neural networks, our approach can easily find interpretable decision rule sets that have comparable predictive performance.