End-to-end Generative Zero-shot Learning via Few-shot Learning
arXiv:2102.04379
Abstract
Contemporary state-of-the-art approaches to Zero-Shot Learning (ZSL) train generative nets to synthesize examples conditioned on the provided metadata. Thereafter, classifiers are trained on these synthetic data in a supervised manner. In this work, we introduce Z2FSL, an end-to-end generative ZSL framework that uses such an approach as a backbone and feeds its synthesized output to a Few-Shot Learning (FSL) algorithm. The two modules are trained jointly. Z2FSL solves the ZSL problem with a FSL algorithm, reducing, in effect, ZSL to FSL. A wide class of algorithms can be integrated within our framework. Our experimental results show consistent improvement over several baselines. The proposed method, evaluated across standard benchmarks, shows state-of-the-art or competitive performance in ZSL and Generalized ZSL tasks.
arXiv comment12 pages, 3 figures, 6 tables
References in corpus (14)
- Auto-Encoding Variational Bayes
- Language Models are Few-Shot Learners
- Conditional Image Synthesis With Auxiliary Classifier GANs
- Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation
- Improved Training of Wasserstein GANs
- CheXNet: Radiologist-Level Pneumonia Detection on Chest X-Rays with Deep Learning
- Matching Networks for One Shot Learning
- Data Augmentation Generative Adversarial Networks
- Learning to Predict 3D Objects with an Interpolation-based Differentiable Renderer
- Unsupervised Learning of Depth and Ego-Motion from Video
- Leveraging the Invariant Side of Generative Zero-Shot Learning
- Latent Embedding Feedback and Discriminative Features for Zero-Shot Classification
- Recovering the Missing Link: Predicting Class-Attribute Associations for Unsupervised Zero-Shot Learning
- Feature Generating Networks for Zero-Shot Learning