arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2209.08448cs.LG
收藏

NeuCEPT:通过精确保证的关键神经元识别局部发现神经网络机理

NeuCEPT: Locally Discover Neural Networks' Mechanism via Critical Neurons Identification with Precision Guarantee

University of Florida(佛罗里达大学)

Minh N. Vu, Truc D. Nguyen, My T. Thai

摘要:

尽管最近有关于理解深度神经网络(DNN)的研究,但关于DNN如何生成预测仍然存在许多问题。特别是,考虑到对不同输入样本的相似预测,生成这些预测的底层机制是否相同?在这项工作中,我们提出了NeuCEPT,一种方法来本地发现关键神经元,在模型的预测中发挥了重要作用,并确定模型的机制,在产生这些预测。我们首先制定了一个关键的神经元识别问题最大化的互信息目标序列,并提供了一个理论框架,有效地解决关键神经元,同时保持精度可控。NeuCEPT接下来以无监督的方式学习不同模型的机制。我们的实验结果表明,NeuCEPT识别的神经元不仅对模型的预测有很强的影响,而且还包含有关模型机制的有意义的信息。

英文摘要:

Despite recent studies on understanding deep neural networks (DNNs), there exists numerous questions on how DNNs generate their predictions. Especially, given similar predictions on different input samples, are the underlying mechanisms generating those predictions the same? In this work, we propose NeuCEPT, a method to locally discover critical neurons that play a major role in the model's predictions and identify model's mechanisms in generating those predictions. We first formulate a critical neurons identification problem as maximizing a sequence of mutual-information objectives and provide a theoretical framework to efficiently solve for critical neurons while keeping the precision under control. NeuCEPT next heuristically learns different model's mechanisms in an unsupervised manner. Our experimental results show that neurons identified by NeuCEPT not only have strong influence on the model's predictions but also hold meaningful information about model's mechanisms.

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。

↑