Unsupervised Meta-Learning through Latent-Space Interpolation in Generative Models

2020/06/18 by Khodadadeh, Siavash, Zehtabian, Sharare, Vahidian, Saeed +3 · 1 citation
#FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · doi:10.48550/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.

Cited by

• Is Meta-Learning Out? Rethinking Unsupervised Few-Shot Classification with Limited Entropy

Related

• Revisiting Meta-Learning as Supervised Learning
• Unsupervised Learning via Meta-Learning
• A Comprehensive Overview and Survey of Recent Advances in Meta-Learning
• Set-based Meta-Interpolation for Few-Task Meta-Learning
• Unsupervised Representation Learning to Aid Semi-Supervised Meta Learning