Meta-Learning for Semi-Supervised Few-Shot Classification

Ren, Mengye; Triantafillou, Eleni; Ravi, S.; Snell, Jake; Swersky, Kevin; Tenenbaum, Joshua B.; Larochelle, Hugo; Zemel, Richard S.

doi:10.48550/arxiv.1803.00676

Cited by 113 publications

(154 citation statements)

References 15 publications

Supporting

Mentioning

151

Contrasting

Order By: Relevance

“…Based on this, however, the model of each part of the region would be vulnerable to very few labeled samples. To solve this problem, we introduced meta-learning technology for few-shot adaption (Ren et al, 2018). An overview of the proposed method is presented in Fig.…”

Section: Overview Of the Methodsmentioning

confidence: 99%

Meta-learning an Intermediate Representation for Few-shot Block-wise Prediction of Landslide Susceptibility

Chen¹,

Ding²,

Hu³

et al. 2021

Preprint

View full text Add to dashboard Cite

Predicting a landslide susceptibility map (LSM) is essential for risk recognition and disaster prevention. Despite the successful application of data-driven prediction approaches, current data-driven methods generally apply a single global model to predict the LSM for an entire target region. However, we argue that, in complex circumstances, especially in large-scale areas, each part of the region holds different landslide-inducing environments, and therefore, should be predicted individually with respective models. In this study, target scenarios were segmented into blocks for individual analysis using topographical factors. But simply conducting training and testing using limited samples within each block is hardly possible for a satisfactory LSM prediction, due to the adverse effect of overfitting. To solve the problems, we train an intermediate representation by the meta-learning paradigm, which is superior for capturing information from LSM tasks in order to generalize proficiently. We chose this based on the hypothesis that there are more general concepts among LSM tasks that are sensitive to variations in input features. Thus, using the intermediate representation, we can easily adapt the model for different blocks or even unseen tasks using few exemplar samples. Experimental results on two study areas demonstrated the validity of our block-wise analysis in large scenarios and revealed the top fewshot adaption performances of the proposed methods. The code will be made publicly available on the website 1 .

show abstract

Section: Overview Of the Methodsmentioning

confidence: 99%

Meta-learning an Intermediate Representation for Few-shot Block-wise Prediction of Landslide Susceptibility

Chen¹,

Ding²,

Hu³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…TieredImageNet dataset is larger in size than miniImageNet, containing 608 classes from ImageNet [21]. Its classes are acquired based on 34 higher-level categories.…”

Section: Experiments Settingmentioning

confidence: 99%

Ontology-based n-ball Concept Embeddings Informing Few-shot Image Classification

Jayathilaka,

Mu,

Sattler

2021

Preprint

View full text Add to dashboard Cite

We propose a novel framework named ViOCE that integrates ontology-based background knowledge in the form of n-ball concept embeddings into a neural network based vision architecture. The approach consists of two components -converting symbolic knowledge of an ontology into continuous space by learning n-ball embeddings that capture properties of subsumption and disjointness, and guiding the training and inference of a vision model using the learnt embeddings. We evaluate ViOCE using the task of few-shot image classification, where it demonstrates superior performance on two standard benchmarks.

show abstract

“…In this section, we present our experimental results in various few-shot learning benchmarks, including miniImageNet (Vinyals et al, 2016), tieredImageNet (Ren et al, 2018), CIFAR-FS (Bertinetto et al, 2018), andFC-100 (Oreshkin et al, 2018) 3 . The miniImageNet dataset consists of 100 classes, sampled from ImageNet (Russakovsky et al, 2015), and randomly split into 64, 16, and 20 classes for training, validation, and testing, respectively.…”

Section: Experimental Studymentioning

confidence: 99%

Task Affinity with Maximum Bipartite Matching in Few-Shot Learning

Le¹,

Dong²,

Soltani³

et al. 2021

Preprint

View full text Add to dashboard Cite

We propose an asymmetric affinity score for representing the complexity of utilizing the knowledge of one task for learning another one. Our method is based on the maximum bipartite matching algorithm and utilizes the Fisher Information matrix. We provide theoretical analyses demonstrating that the proposed score is mathematically well-defined, and subsequently use the affinity score to propose a novel algorithm for the few-shot learning problem. In particular, using this score, we find relevant training data labels to the test data and leverage the discovered relevant data for episodically fine-tuning a few-shot model. Results on various few-shot benchmark datasets demonstrate the efficacy of the proposed approach by improving the classification accuracy over the state-of-the-art methods even when using smaller models.

show abstract

Meta-Learning for Semi-Supervised Few-Shot Classification

Cited by 113 publications

References 15 publications

Meta-learning an Intermediate Representation for Few-shot Block-wise Prediction of Landslide Susceptibility

Meta-learning an Intermediate Representation for Few-shot Block-wise Prediction of Landslide Susceptibility

Ontology-based n-ball Concept Embeddings Informing Few-shot Image Classification

Task Affinity with Maximum Bipartite Matching in Few-Shot Learning

Contact Info

Product

Resources

About