Unsupervised Meta-Learning through Latent-Space Interpolation in Generative Models
arXiv:2006.10236
Abstract
Unsupervised meta-learning approaches rely on synthetic meta-tasks that are created using techniques such as random selection, clustering and/or augmentation. Unfortunately, clustering and augmentation are domain-dependent, and thus they require either manual tweaking or expensive learning. In this work, we describe an approach that generates meta-tasks using generative models. A critical component is a novel approach of sampling from the latent space that generates objects grouped into synthetic classes forming the training and validation data of a meta-task. We find that the proposed approach, LAtent Space Interpolation Unsupervised Meta-learning (LASIUM), outperforms or is competitive with current unsupervised learning baselines on few-shot classification tasks on the most widely used benchmark datasets. In addition, the approach promises to be applicable without manual tweaking over a wider range of domains than previous approaches.
References in corpus (12)
- Auto-Encoding Variational Bayes
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- Prototypical Networks for Few-shot Learning
- Theoretical Models of Learning to Learn
- A Simple Neural Attentive Meta-Learner
- Adversarially Learned Inference
- Learning to Propagate Labels: Transductive Propagation Network for Few-shot Learning
- On First-Order Meta-Learning Algorithms
- Meta-Learning with Implicit Gradients
- Unsupervised Meta-Learning For Few-Shot Image Classification
- Understanding and Improving Interpolation in Autoencoders via an Adversarial Regularizer
- Assume, Augment and Learn: Unsupervised Few-Shot Meta-Learning via Random Labels and Data Augmentation