Abstract: This paper focuses on detection tasks in information extraction, where positive instances are sparsely distributed and models are usually evaluated using F-measure on positive classes. These characteristics often result in deficient per-formance of neural network based detection models. In this paper, we propose adaptive scaling, an algorithm which can handle the positive sparsity problem and directly optimize over F-measure via dynamic cost-sensitive learning. To this end, we borrow the idea of marginal utility from economics and propose a theoretical framework for instance importance measuring without introducing any additional hyper-parameters. Experiments show that our algorithm leads to a more effective and stable training of neural network based detection models.
Authors: Hongyu Lin, Yaojie Lu, Xianpei Han, Le Sun (University of Chinese Academy of Sciences)