2017/03/19 by Florian Dumpert, Dumpert, Florian
Computer Science · #62G08 #62G20 #62G35 #FOS: Computer and information sciences #Face and Expression Recognition #Machine Learning (stat.ML) #Machine Learning and Algorithms #Machine Learning and Data Classification #Machine Learning and ELM
paper · pdf · doi:10.48550/arxiv.1703.06528
openalex publication_date 2017/03/19 · openalex created_date 2022/10/04 · openalex updated_date 2026/07/28
The massive amount of available data potentially used to discover patters in\nmachine learning is a challenge for kernel based algorithms with respect to\nruntime and storage capacities. Local approaches might help to relieve these\nissues. From a statistical point of view local approaches allow additionally to\ndeal with different structures in the data in different ways. This paper\nanalyses properties of localized kernel based, non-parametric statistical\nmachine learning methods, in particular of support vector machines (SVMs) and\nmethods close to them. We will show there that locally learnt kernel methods\nare universal consistent. Furthermore, we give an upper bound for the maxbias\nin order to show statistical robustness of the proposed method.\n