arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2303.09154stat.MLcs.AIcs.LGmath.STstat.TH
收藏

多任务线性神经网络的贝叶斯泛化误差

Bayesian Generalization Error in Linear Neural Networks with Concept Bottleneck Structure and Multitask Formulation

Aisin Corporation(爱信公司)

Naoki Hayashi, Yoshihide Sawada

摘要:

概念瓶颈模型(CBM)是一种普遍存在的用概念解释神经网络的方法。在CBM中,概念作为可观察值插入到输出层和最后一个中间层之间。这有助于理解神经网络生成输出背后的原因:权重对应于从最后一个隐藏层到输出层的概念。然而,由于神经网络一般是一个奇异的统计模型,因此还不可能理解CBM中泛化误差的行为。当模型为奇异时,无法创建从参数到概率分布的一对一映射。这种不可识别性使得分析泛化性能变得困难。在这项研究中,我们从数学上澄清了贝叶斯泛化误差和自由能的CBM时,其结构是三层线性神经网络。我们还考虑了一个多任务的问题,神经网络的输出不仅是原始的输出,但也的概念。结果表明,与标准版本相比,CBM极大地改变了三层线性神经网络中参数区域的行为和贝叶斯泛化误差,而多任务模型则没有.

英文摘要:

Concept bottleneck model (CBM) is a ubiquitous method that can interpret neural networks using concepts. In CBM, concepts are inserted between the output layer and the last intermediate layer as observable values. This helps in understanding the reason behind the outputs generated by the neural networks: the weights corresponding to the concepts from the last hidden layer to the output layer. However, it has not yet been possible to understand the behavior of the generalization error in CBM since a neural network is a singular statistical model in general. When the model is singular, a one to one map from the parameters to probability distributions cannot be created. This non-identifiability makes it difficult to analyze the generalization performance. In this study, we mathematically clarify the Bayesian generalization error and free energy of CBM when its architecture is three-layered linear neural networks. We also consider a multitask problem where the neural network outputs not only the original output but also the concepts. The results show that CBM drastically changes the behavior of the parameter region and the Bayesian generalization error in three-layered linear neural networks as compared with the standard version, whereas the multitask formulation does not.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑