Unsupervised Meta-Learning For Few-Shot Image Classification

2018/11/28 by Siavash Khodadadeh, Khodadadeh, Siavash, Ladislau Bölöni +3 · 65 citations
Biochemistry, Genetics and Molecular Biology · Computer Science · #Artificial intelligence #Artificial neural network #Cancer-related molecular mechanisms research #Class (philosophy) #Computer Vision and Pattern Recognition (cs.CV) #Computer science #Contextual image classification #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Image (mathematics) #Machine Learning (cs.LG) #Machine learning #Meta learning (computer science) #Multi-task learning #Multimodal Machine Learning Applications #Pattern recognition (psychology) #Semi-supervised learning #Set (abstract data type) #Supervised learning #Task (project management) #Unsupervised learning #cs.CV #cs.LG

paper · pdf · doi:10.48550/arxiv.1811.11819

published in arXiv (Cornell University) 32, 10132-10142 (Cornell University)

openalex publication_date 2018/11/28 · arxiv created 2019/11/07 · arxiv updated 2019/11/11 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/06

Abstract

Few-shot or one-shot learning of classifiers requires a significant inductive bias towards the type of task to be learned. One way to acquire this is by meta-learning on tasks similar to the target task. In this paper, we propose UMTRA, an algorithm that performs unsupervised, model-agnostic meta-learning for classification tasks. The meta-learning step of UMTRA is performed on a flat collection of unlabeled images. While we assume that these images can be grouped into a diverse set of classes and are relevant to the target task, no explicit information about the classes or any labels are needed. UMTRA uses random sampling and augmentation to create synthetic training tasks for meta-learning phase. Labels are only needed at the final target task learning step, and they can be as little as one sample per class. On the Omniglot and Mini-Imagenet few-shot learning benchmarks, UMTRA outperforms every tested approach based on unsupervised learning of representations, while alternating for the best performance with the recent CACTUs algorithm. Compared to supervised model-agnostic meta-learning approaches, UMTRA trades off some classification accuracy for a reduction in the required labels of several orders of magnitude.

Cited by

• Heterogeneous Adversarial Play in Interactive Environments
• Directed-MAML: Meta Reinforcement Learning Algorithm with Task-directed Approximation
• Is Meta-Learning Out? Rethinking Unsupervised Few-Shot Classification with Limited Entropy
• Divergent Search for Few-Shot Image Classification
• Are Fewer Labels Possible for Few-shot Learning?
• Unsupervised Ground Metric Learning
• Few-Shot Unsupervised Continual Learning through Meta-Examples
• Mask6D: Masked Pose Priors For 6D Object Pose Estimation
• Trainable Class Prototypes for Few-Shot Learning
• Meta-learning of Pooling Layers for Character Recognition
• ProtoTransformer: A Meta-Learning Approach to Providing Student Feedback
• Online Unsupervised Learning of Visual Representations and Categories
• Meta-Learning Requires Meta-Augmentation
• Semi-supervised Learning with a Teacher-student Network for Generalized Attribute Prediction
• iADCPS: Time Series Anomaly Detection for Evolving Cyber-physical Systems via Incremental Meta-learning

Related

• Revisiting Unsupervised Meta-Learning via the Characteristics of Few-Shot Tasks
• MetaDelta: A Meta-Learning System for Few-shot Image Classification
• Rethinking Few-Shot Image Classification: a Good Embedding Is All You Need?
• Learning from Few Samples: A Survey
• Unsupervised Learning via Meta-Learning