Comparing Transfer and Meta Learning Approaches on a Unified Few-Shot Classification Benchmark

2021-04-06 · Vincent Dumoulin, Neil Houlsby, Utku Evci, Xiaohua Zhai, Ross Goroshin, Sylvain Gelly, Hugo Larochelle

Meta and transfer learning are two successful families of approaches to few-shot learning. Despite highly related goals, state-of-the-art advances in each family are measured largely in isolation of each other. As a result of diverging evaluation norms, a direct or thorough comparison of different approaches is challenging. To bridge this gap, we perform a cross-family study of the best transfer and meta learners on both a large-scale meta-learning benchmark (Meta-Dataset, MD), and a transfer learning benchmark (Visual Task Adaptation Benchmark, VTAB). We find that, on average, large-scale transfer methods (Big Transfer, BiT) outperform competing approaches on MD, even when trained only on ImageNet. In contrast, meta-learning approaches struggle to compete on VTAB when trained and validated on MD. However, BiT is not without limitations, and pushing for scale does not improve performance on highly out-of-distribution MD tasks. In performing this study, we reveal a number of discrepancies in evaluation norms and study some of these in light of the performance gap. We hope that this work facilitates sharing of insights from each community, and accelerates progress on few-shot learning.

📄 PDF Abstract BibTeX arXiv:2104.02638

Code (1)

google-research/meta-dataset 공식 구현 tf

Tasks

Few-Shot Learning General Classification Meta-Learning Transfer Learning

A Unified Framework with Meta-dropout for Few-shot Learning

2022-10-12 · Shaobo Lin, Xingyu Zeng, Rui Zhao

Conventional training of deep neural networks usually requires a substantial amount of data with expensive human annotations. In this paper, we utilize the idea of meta-learning to explain two very different streams of f…

Few-Shot Image Classification Few-Shot Learning Few-Shot Object Detection image-classification+4

MDFL: A UNIFIED FRAMEWORK WITH META-DROPOUT FOR FEW-SHOT LEARNING

2021-09-29 · Shaobo Lin, Xingyu Zeng, Rui Zhao

Few-Shot Image Classification Few-Shot Learning Few-Shot Object Detection image-classification+4

The Curse of Low Task Diversity: On the Failure of Transfer Learning to Outperform MAML and Their Empirical Equivalence

2022-08-02 · Brando Miranda, Patrick Yu, Yu-Xiong Wang, Sanmi Koyejo

Recently, it has been observed that a transfer learning solution might be all we need to solve many few-shot learning benchmarks -- thus raising important questions about when and how meta-learning algorithms should be d…

Diversity Few-Shot Learning Meta-Learning Transfer Learning

A Meta-Learning Approach for Custom Model Training

2018-09-21 · Amir Erfan Eshratifar, Mohammad Saeed Abrishami, David Eigen, Massoud Pedram

Transfer-learning and meta-learning are two effective methods to apply knowledge learned from large data sources to new tasks. In few-class, few-shot target task settings (i.e. when there are only a few classes and train…

Meta-Learning model Transfer Learning

Self-Supervised Prototypical Transfer Learning for Few-Shot Classification

2020-06-19 · Carlos Medina, Arnout Devos, Matthias Grossglauser

Most approaches in few-shot learning rely on costly annotated data related to the goal task domain during (pre-)training. Recently, unsupervised meta-learning methods have exchanged the annotation requirement for a reduc…

Classification Few-Shot Learning General Classification Meta-Learning+4