Tabular machine learning using conjunctive threshold neural networks
Tabular machine learning using conjunctive threshold neural networks
复制标题
DOI：
10.1016/j.mlwa.2022.100429
复制
发表时间：
2022-10
影响因子：
--
通讯作者：
Weijia Wang;Litao Qiao;Bill Lin
中科院分区：
文献类型：
--
作者：
Weijia Wang;Litao Qiao;Bill Lin
我们提出了一种新颖的三层神经网络架构，具有针对表格数据分类问题的阈值激活。隐藏层单元对应于具有任意权重和偏差以及步骤激活的可训练神经元。这些神经元在逻辑上相当于阈值逻辑函数。输出层神经元也是一个阈值函数，实现隐藏层阈值函数的结合。这种神经网络架构可以利用最先进的网络训练方法来实现高预测精度，并且网络的设计使得可以很容易地从模型中得出最少的人类可理解的解释。此外，我们采用稀疏促进正则化方法来稀疏阈值函数以简化它们，并稀疏输出神经元，使其仅依赖于隐藏层阈值函数的一小部分。实验结果表明，我们的方法在预测准确性方面优于其他最先进的可解释决策模型。
We propose a novel three-layer neural network architecture with threshold activations for tabular data classification problems. The hidden layer units correspond to trainable neurons with arbitrary weights and biases and a step activation. These neurons are logically equivalent to threshold logic functions. The output layer neuron is also a threshold function that implements a conjunction of the hidden layer threshold functions. This neural network architecture can leverage state-of-the-art network training methods to achieve high prediction accuracy, and the network is designed so that minimal human understandable explanations can be readily derived from the model. Further, we employ a sparsity-promoting regularization approach to sparsify the threshold functions to simplify them, and to sparsify the output neuron so that it only depends on a small subset of hidden layer threshold functions. Experimental results show that our approach outperforms other state-of-the-art interpretable decision models in prediction accuracy.