首页|Data on Machine Learning Reported by Researchers at University of Patras (Adapti ve Augmentation Framework for Domain Independent Few Shot Learning A)
Data on Machine Learning Reported by Researchers at University of Patras (Adapti ve Augmentation Framework for Domain Independent Few Shot Learning A)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – New research on Machine Learning is th e subject of a report. According to news reporting from Patras, Greece, by NewsR x journalists, research stated, “Few-Shot learning is a research area of machine learning, which aims to develop a prediction model based on a limited set of tr aining instances. In contrast to human learners, who are able to quickly learn a nd adapt to new tasks, machine learning models require large amounts of training instances in order to generalize efficiently.” The news correspondents obtained a quote from the research from the University o f Patras, “Image augmentation provides a potential solution to this challenge in few-shot learning by enlarging the training dataset. However, an excessive and uncontrollable enlargement of the initial dataset may potentially add noise, whi ch could significantly impair the learning efficacy, especially in the few-shot context. Our motivation lies in the fact that since the least confident instance s are the hardest to classify, performing targeted augmentations on these instan ces could efficiently enhance the limited representational sample space in a few shot context. In this work, we propose a new augmentation-based prediction frame work, which adaptively enlarges the few-shot training samples by performing targ eted image augmentations for the hardest to identify instances. Given the inhere ntly limited size of data, their proper identification is challenging. Therefore , we adopt a Least Confident Augmentation strategy based on the output confidenc es of an embedding-based estimator. In addition, we introduce an adaptive cleani ng step in order to remove the potential noise added during the targeted augment ations.”
PatrasGreeceEuropeCyborgsEmergin g TechnologiesMachine LearningUniversity of Patras
NETL
NSTL
