Sketch In, Sketch Out: Accelerating both Learning and Inference for Structured Prediction with Kernels

arXiv:2302.10128v2 Announce Type: replace-cross
Abstract: Leveraging the kernel trick in both the input and output spaces, surrogate kernel methods are a flexible and theoretically grounded solution to structured output prediction. If they provide state-of-the-art performance on complex data sets of moderate size (e.g., in chemoinformatics), these approaches however fail to scale. We propose to equip surrogate kernel methods with sketching-based approximations, applied to both the input and output feature maps. We prove excess risk bounds on the original structured prediction …

abstract art arxiv cs.lg data data sets inference kernel performance prediction solution spaces state stat.ml trick type