We propose a novel class of deep stochastic predictors for classifying metric data on graphs within the PAC-Bayes risk certification paradigm. Classifiers are realized as linearly parametrized deep assignment flows with random initial conditions. Building on the recent PAC-Bayes literature and data-dependent priors, this approach enables (i) to use risk bounds as training objectives for learning posterior distributions on the hypothesis space and (ii) to compute tight out-of-sample risk certificates of randomized classifiers more efficiently than related work. Comparison with empirical test set errors illustrates the performance and practicality of this self-certifying classification method.
翻译:我们建议建立一个新型的深随机预测器类别,用于在PAC-Bayes风险认证范式内对图表上的计量数据进行分类,分类器作为线性平衡的深度派任流动和随机初始条件实现,以最近的PAC-Bayes文献和数据依赖前科为基础,这种方法使(一) 利用风险界限作为培训目标,在假设空间上学习后方分布,(二) 以比相关工作更有效率的方式计算随机派任人员严格、不具有抽样风险的证书,与经验测试组的错误进行比较,说明这种自证分类方法的性能和实用性。