Symbolic techniques based on Satisfiability Modulo Theory (SMT) solvers have been proposed for analyzing and verifying neural network properties, but their usage has been fairly limited owing to their poor scalability with larger networks. In this work, we propose a technique for combining gradient-based methods with symbolic techniques to scale such analyses and demonstrate its application for model explanation. In particular, we apply this technique to identify minimal regions in an input that are most relevant for a neural network's prediction. Our approach uses gradient information (based on Integrated Gradients) to focus on a subset of neurons in the first layer, which allows our technique to scale to large networks. The corresponding SMT constraints encode the minimal input mask discovery problem such that after masking the input, the activations of the selected neurons are still above a threshold. After solving for the minimal masks, our approach scores the mask regions to generate a relative ordering of the features within the mask. This produces a saliency map which explains "where a model is looking" when making a prediction. We evaluate our technique on three datasets - MNIST, ImageNet, and Beer Reviews, and demonstrate both quantitatively and qualitatively that the regions generated by our approach are sparser and achieve higher saliency scores compared to the gradient-based methods alone. Code and examples are at - https://github.com/google-research/google-research/tree/master/smug_saliency
翻译:以满足性要求的 Modulo Theory (SMT) 解析和核查神经网络特性的方法提出了基于满足性可满足性 Modulo Theory (SMT) 的符号技术,但是由于神经网络特性的伸缩性差,它们的使用相当有限。 在这项工作中,我们提出了一种方法,将基于梯度的方法与象征性技术相结合,以扩大这种分析的规模,并展示其用于示范解释的应用。特别是,我们运用这种技术,在与神经网络预测最相关的投入中确定最低区域。我们的方法使用梯度信息(以综合梯度为基础),以第一层的一组神经元为重点,使我们的技术能够推广到大型网络。相应的SMTM限制将最小的输入掩码发现问题编码编码化为最小,在隐藏输入后,选定神经神经元的激活率仍然高于阈值。我们的方法在解决了最起码的面具后,将遮盖区域评分出一个相对排序。这产生了一个显著的地图,用来解释在作出预测时“模型正在寻找的” 。 我们评估三个数据集的技术- MNIST、 imNet- 和Bregrales- 和Begralearger- exerview as real asyal as as as as asyal asyal asyal asureal as as as as as as as as asilus asilus as as as asilus asilus