论文标题
使用潜在几何图中的标签变化对深度学习概括进行排名
Ranking Deep Learning Generalization using Label Variation in Latent Geometry Graphs
论文作者
论文摘要
在不依赖验证集的情况下测量深神经网络（DNN）的概括性能是一项艰巨的任务。在这项工作中，我们建议利用潜在几何图（LGG）代表训练有素的DNN体系结构的潜在空间。通过连接在所考虑的DNN的给定层上产生相似的潜在表示的样品来获得此类图。然后，我们通过查看与LGG中不同类别的样本的紧密连接来获得概括得分。这个分数使我们能够在2020年神经2020中排名第三，以预测深度学习（PGDL）竞争的概括。
Measuring the generalization performance of a Deep Neural Network (DNN) without relying on a validation set is a difficult task. In this work, we propose exploiting Latent Geometry Graphs (LGGs) to represent the latent spaces of trained DNN architectures. Such graphs are obtained by connecting samples that yield similar latent representations at a given layer of the considered DNN. We then obtain a generalization score by looking at how strongly connected are samples of distinct classes in LGGs. This score allowed us to rank 3rd on the NeurIPS 2020 Predicting Generalization in Deep Learning (PGDL) competition.
