Meta-Learned Attribute Self-Gating for Continual Generalized Zero-Shot Learning

Abstract

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 generative models to combat bias towards classes seen during training have pushed the state of the art of ZSL, but these generative models can be slow or computationally expensive to train. Additionally, while many previous ZSL methods assume a one-time adaptation to unseen classes, in reality, the world is always changing, necessitating a constant adjustment for deployed models. Models unprepared to handle a sequential stream of data are likely to experience catastrophic forgetting. We propose a meta-continual zero-shot learning (MCZSL) approach to address both these issues. In particular, by pairing self-gating of attributes and scaled class normalization with meta-learning based training, we are able to outperform state-of-the-art results while being able to train our models substantially faster ($>100\times$) than expensive generative-based approaches. We demonstrate this by performing experiments on five standard ZSL datasets (CUB, aPY, AWA1, AWA2 and SUN) in both generalized zero-shot learning and generalized continual zero-shot learning settings.

Under Review

References in corpus (8)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Efficient Lifelong Learning with A-GEM
• Less-forgetting Learning in Deep Neural Networks
• A Unified Semantic Embedding: Relating Taxonomies and Attributes
• Zero-shot Learning via Simultaneous Generating and Learning
• Learning Joint Feature Adaptation for Zero-Shot Recognition
• Isometric Propagation Network for Generalized Zero-shot Learning
• Generalized Continual Zero-Shot Learning