2012/02/28 by Emilie Morvant, Morvant, Emilie, Sokol Koço +3
Computer Science · #Adversarial Robustness in Machine Learning #FOS: Computer and information sciences #Face and Expression Recognition #Imbalanced Data Classification Techniques #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Machine Learning and Algorithms #Stochastic Gradient Optimization Techniques
paper · pdf · doi:10.48550/arxiv.1202.6228
openalex publication_date 2012/02/28 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
In this work, we propose a PAC-Bayes bound for the generalization risk of the\nGibbs classifier in the multi-class classification framework. The novelty of\nour work is the critical use of the confusion matrix of a classifier as an\nerror measure; this puts our contribution in the line of work aiming at dealing\nwith performance measure that are richer than mere scalar criterion such as the\nmisclassification rate. Thanks to very recent and beautiful results on matrix\nconcentration inequalities, we derive two bounds showing that the true\nconfusion risk of the Gibbs classifier is upper-bounded by its empirical risk\nplus a term depending on the number of training examples in each class. To the\nbest of our knowledge, this is the first PAC-Bayes bounds based on confusion\nmatrices.\n