From generalized zero-shot learning to long-tail with class descriptors

Samuel, Dvir; Atzmon, Yuval; Chechik, Gal

doi:10.1109/wacv48630.2021.00033

Cited by 28 publications

(14 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…GZSL also can be combined with reinforcement learning [197], [210], [211] to better tackle new tasks. In addition, GZSL can be extended in several fronts, which include multimodal learning [85], [199], [200], multi-label learning [190], multi-view learning [191], weakly supervised learning to progressively incorporate training instances from easy to hard [100], [130], [212], continual learning [110], long-tail learning [213], or online learning for few-shot learning where a small portion of labeled samples from some classes are available [67].…”

Section: Research Gapsmentioning

confidence: 99%

A Review of Generalized Zero-Shot Learning Methods

Pourpanah

Abdar

Luo

et al. 2022

IEEE Trans. Pattern Anal. Mach. Intell.

166

View full text Add to dashboard Cite

Generalized zero-shot learning (GZSL) aims to train a model for classifying data samples under the condition that some output classes are unknown during supervised learning. To address this challenging task, GZSL leverages semantic information of the seen (source) and unseen (target) classes to bridge the gap between both seen and unseen classes. Since its introduction, many GZSL models have been formulated. In this review paper, we present a comprehensive review on GZSL. Firstly, we provide an overview of GZSL including the problems and challenges. Then, we introduce a hierarchical categorization for the GZSL methods and discuss the representative methods in each category. In addition, we discuss the available benchmark data sets and applications of GZSL, along with a discussion on the research gaps and directions for future investigations.

show abstract

Section: Research Gapsmentioning

confidence: 99%

A Review of Generalized Zero-Shot Learning Methods

Pourpanah

Abdar

Luo

et al. 2022

IEEE Trans. Pattern Anal. Mach. Intell.

166

View full text Add to dashboard Cite

show abstract

“…Recently, Samuel et al [28] leveraged class descriptors to facilitate long-tailed classification. It developed a dual network to derive both visual features and semantic features from the input image, and then fused these two together to boost the performance of long-tailed classification.…”

Section: Related Workmentioning

confidence: 99%

Tackling Long-Tailed Category Distribution Under Domain Shifts

Gu¹,

Guo²,

Li³

et al. 2022

Preprint

View full text Add to dashboard Cite

Machine learning models fail to perform well on real-world applications when 1) the category distribution P (Y ) of the training dataset suffers from long-tailed distribution and 2) the test data is drawn from different conditional distributions P (X|Y ). Existing approaches cannot handle the scenario where both issues exist, which however is common for real-world applications. In this study, we took a step forward and looked into the problem of long-tailed classification under domain shifts. We designed three novel core functional blocks including Distribution Calibrated Classification Loss, Visual-Semantic Mapping and Semantic-Similarity Guided Augmentation. Furthermore, we adopted a meta-learning framework which integrates these three blocks to improve domain generalization on unseen target domains. Two new datasets were proposed for this problem, named AWA2-LTS and ImageNet-LTS. We evaluated our method on the two datasets and extensive experimental results demonstrate that our proposed method can achieve superior performance over state-of-the-art long-tailed/domain generalization approaches and the combinations. Source codes and datasets can be found at our project page https://xiaogu.site/LTDS.

show abstract

“…No label is used for the architecture search. In Table 1, we summarize the results of our experiments and compare them against the previous representative works on long-tailed distributions: ResNet-32 + Focal loss (Lin et al, 2017), ResNet-32 + Sigmoid (SGM) and Balanced Sigmoid (BSGM) Cross Entropy losses (Cui et al, 2019), LDAM-DRW (Cao et al, 2019), smDragon and VE2 + smDragon (Samuel et al, 2021), SSNAS (Kaplan & Giryes, 2020). To provide a fair comparison to SSNAS method, we implement it in a common framework with common hyper-parameters.…”

Section: Evaluation On a Long-tailed Distributionmentioning

confidence: 99%

Self-Supervised Neural Architecture Search for Imbalanced Datasets

Тимофеев¹,

Chrysos²,

Cevher³

2021

Preprint

View full text Add to dashboard Cite

Neural Architecture Search (NAS) provides stateof-the-art results when trained on well-curated datasets with annotated labels. However, annotating data or even having balanced number of samples can be a luxury for practitioners from different scientific fields, e.g., in the medical domain.To that end, we propose a NAS-based framework that bears the threefold contributions: (a) we focus on the self-supervised scenario, i.e., where no labels are required to determine the architecture, and (b) we assume the datasets are imbalanced, (c) we design each component to be able to run on a resource constrained setup, i.e., on a single GPU (e.g. Google Colab). Our components build on top of recent developments in self-supervised learning (Zbontar et al., 2021), self-supervised NAS (Kaplan & Giryes, 2020) and extend them for the case of imbalanced datasets. We conduct experiments on an (artificially) imbalanced version of CIFAR-10 and we demonstrate our proposed method outperforms standard neural networks, while using 27× less parameters. To validate our assumption on a naturally imbalanced dataset, we also conduct experiments on ChestM-NIST and COVID-19 X-ray. The results demonstrate how the proposed method can be used in imbalanced datasets, while it can be fully run on a single GPU. Code is available here. Preliminaries2.1. Neural Architecture Search Neural Architecture Search (NAS) can be roughly separated into three components (Elsken et al., 2019): search space, search strategy, and performance estimation strategy. The first component defines the set of architectures to be

show abstract

From generalized zero-shot learning to long-tail with class descriptors

Cited by 28 publications

References 30 publications

A Review of Generalized Zero-Shot Learning Methods

A Review of Generalized Zero-Shot Learning Methods

Tackling Long-Tailed Category Distribution Under Domain Shifts

Self-Supervised Neural Architecture Search for Imbalanced Datasets

Contact Info

Product

Resources

About