Class Normalization for (Continual)? Generalized Zero-Shot Learning
Normalization techniques have proved to be a crucial ingredient of successful training in a traditional supervised learning regime. However, in the zero-shot learning (ZSL) world, these ideas have received only marginal attention. This work studies normalization in ZSL scenario from both theoretical and practical perspectives. First, we give a theoretical explanation to two popular tricks used in zero-shot learning: normalize+scale and attributes normalization and show that they help training by preserving variance during a forward pass. Next, we demonstrate that they are insufficient to normalize a deep ZSL model and propose Class Normalization (CN): a normalization scheme, which alleviates this issue both provably and in practice. Third, we show that ZSL models typically have more irregular loss surface compared to traditional classifiers and that the proposed method partially remedies this problem. Then, we test our approach on 4 standard ZSL datasets and outperform sophisticated modern SotA with a simple MLP optimized without any bells and whistles and having ~50 times faster training speed. Finally, we generalize ZSL to a broader problem -- continual ZSL, and introduce some principled metrics and rigorous baselines for this new setup. The project page is located at https://universome.github.io/class-norm.
Code (3)
Tasks
Generalized Zero-Shot LearningZero-Shot LearningSimilar Papers 제목 키워드 기반
Meta-Learned Attribute Self-Gating for Continual Generalized Zero-Shot Learning
Zero-shot learning (ZSL) has been shown to be a promising approach to generalizing a model to categories unseen during training by leveraging class attributes, but challenges still remain. Recently, methods using generat…
AttributeGeneralized Zero-Shot LearningMeta-LearningZero-Shot LearningContinual Learning Improves Zero-Shot Action Recognition
Zero-shot action recognition requires a strong ability to generalize from pre-training and seen classes to novel unseen classes. Similarly, continual learning aims to develop models that can generalize effectively and le…
Action RecognitionContinual LearningZero-Shot Action RecognitionZero-Shot LearningClass Normalization for Zero-Shot Learning
Normalization techniques have proved to be a crucial ingredient of successful training in a traditional supervised learning regime. However, in zero-shot learning (ZSL) world, these ideas have recieved only marginal atte…
Zero-Shot LearningMeta-Learned Attribute Self-Interaction Network for Continual and Generalized Zero-Shot Learning
Zero-shot learning (ZSL) is a promising approach to generalizing a model to categories unseen during training by leveraging class attributes, but challenges remain. Recently, methods using generative models to combat bia…
AttributeGeneralized Zero-Shot LearningMeta-LearningZero-Shot LearningGeneralized Continual Zero-Shot Learning
Recently, zero-shot learning (ZSL) emerged as an exciting topic and attracted a lot of attention. ZSL aims to classify unseen classes by transferring the knowledge from seen classes to unseen classes based on the class d…
Continual LearningKnowledge DistillationZero-Shot Learning