In zero-shot learning (ZSL), conditional generators have been widely used to generate additional training features. These features can then be used to train the classifiers for testing data. However, some testing data are considered "hard" as they lie close to the decision boundaries and are prone to misclassification, leading to performance degradation for ZSL. In this paper, we propose to learn clusterable features for ZSL problems. Using a Conditional Variational Autoencoder (CVAE) as the feature generator, we project the original features to a new feature space supervised by an auxiliary classification loss. To further increase clusterability, we fine-tune the features using Gaussian similarity loss. The clusterable visual features are not only more suitable for CVAE reconstruction but are also more separable which improves classification accuracy. Moreover, we introduce Gaussian noise to enlarge the intra-class variance of the generated features, which helps to improve the classifier's robustness. Our experiments on SUN,CUB, and AWA2 datasets show consistent improvement over previous state-of-the-art ZSL results by a large margin. In addition to its effectiveness on zero-shot classification, experiments show that our method to increase feature clusterability benefits few-shot learning algorithms as well.
翻译:在零光学习( ZSL) 中, 有条件的生成器被广泛用于生成额外的培训功能。 这些功能随后可用于培训分类人员测试数据。 但是, 一些测试数据被认为“ 硬”, 因为它们靠近决定边界, 容易被错误分类, 导致 ZSL 的性能退化。 在本文中, 我们提议学习ZSL 问题的可分组特性。 使用条件变换自动编码器( CVAE) 来生成功能生成功能, 我们用辅助分类损失所监督的新功能空间来投射原始特性。 为了进一步提高集束性, 我们用高萨相似性损失来微调这些特性。 可分组的视觉特征不仅更适合 CVAE 重建, 并且更具有可变性性, 从而提高分类准确性。 此外, 我们引入高萨噪音来扩大所生成特性的等级内部差异, 从而帮助提高分类人的稳健性。 我们在SUN、 CUB 和 AW2 数据集上进行的实验显示, 相对于先前的状态的分类, 我们用高萨略相似性损失来微的图像模型显示这些特性的特性的特性, 将提高我们的ZSLLAVAVAL 的模型的模型的特性, 。