arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.01284cs.LGcs.CV
收藏

Zero-Shot分类中的性能变化

Performance Variability in Zero-Shot Classification

Universidad Nacional de Córdoba(科尔多瓦国立大学);CONICET(阿根廷国家科学技术研究委员会)

Matías Molina, Jorge Sánchez

摘要:

Zero-shot Classification(ZSC)是学习训练过程中看不到的类的预测器的任务。虽然文献中的不同方法使用相同的类划分进行评估,但对它们在不同类划分下的稳定性知之甚少。在这项工作中,我们通过实验表明,ZSC性能在不断变化的训练设置下表现出很强的可变性。我们建议使用集成学习作为一种尝试,以减轻这种现象。

英文摘要:

Zero-shot classification (ZSC) is the task of learning predictors for classes not seen during training. Although the different methods in the literature are evaluated using the same class splits, little is known about their stability under different class partitions. In this work we show experimentally that ZSC performance exhibits strong variability under changing training setups. We propose the use ensemble learning as an attempt to mitigate this phenomena.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑