Machine learning with limited data
Not on the map. Why?
Abstract (arXiv)
Thanks to the availability of powerful computing resources, big data and deep learning algorithms, we have made great progress on computer vision in the last few years. Computer vision systems begin to surpass humans in some tasks, such as object recognition, object detection, face recognition and pose estimation. Lots of computer vision algorithms have been deployed to real world applications and started to improve our life quality. However, big data and labels are not always available. Sometimes we only have very limited labeled data, such as medical images which requires experts to label them. In this paper, we study few shot image classification, in which we only have very few labeled data. Machine learning with little data is a big challenge. To tackle this challenge, we propose two methods and test their effectiveness thoroughly. One method is to augment image features by mixing the style of these images. The second method is applying spatial attention to explore the relations between patches of images. We also find that domain shift is a critical issue in few shot learning when the training domain and testing domain are different. So we propose a more realistic cross-domain few-shot learning with unlabeled data setting, in which some unlabeled data is available in the target domain. We propose two methods in this setting. Our first method transfers the style information of the unlabeled target dataset to the samples in the source dataset and trains a model with stylized images and original images. Our second method proposes a unified framework to fully utilize all the data. Both of our methods surpass the baseline method by a large margin.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 41 of its 61 references that are in the map. First posted on arXiv 18 Jan 2021.
Builds on
41 of 61 references.
References · 41
- 2017Attention Is All You Needcited by 59,932
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
- 2020Language Models are Few-Shot Learnerscited by 32,199
- 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
- 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
- 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
- 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
- 2017Prototypical Networks for Few-shot Learningcited by 3,001
- 2017Non-local Neural Networkscited by 2,290
- 2016Matching Networks for One Shot Learningcited by 2,246
- 2017Arbitrary Style Transfer in Real-time with Adaptive Instance Normalizationcited by 1,998
29 more
- 2018ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustnesscited by 1,430
- 2018Moment Matching for Multi-Source Domain Adaptationcited by 1,327
- 2017Adversarial Discriminative Domain Adaptationcited by 1,309
- 2017Learning to Compare: Relation Network for Few-Shot Learningcited by 1,139
- 2018On First-Order Meta-Learning Algorithmscited by 1,003
- 2020Meta-Learning in Neural Networks: A Surveycited by 944
- 2018AutoAugment: Learning Augmentation Policies from Datacited by 728
- 2019A Closer Look at Few-shot Classificationcited by 679
- 2019Meta-Learning with Implicit Gradientscited by 523
- 2015A Neural Algorithm of Artistic Stylecited by 480
- 2019Generalizing from a Few Examples: A Survey on Few-Shot Learningcited by 462
- 2019Meta-Learning with Differentiable Convex Optimizationcited by 461
- 2020Rethinking Few-Shot Image Classification: a Good Embedding Is All You Need?cited by 435
- 2018Dynamic Few-Shot Visual Learning without Forgettingcited by 385
- 2019Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAMLcited by 359
- 2016A Learned Representation For Artistic Stylecited by 340
- 2019A Baseline for Few-Shot Image Classificationcited by 298
- 2017Data Augmentation Generative Adversarial Networkscited by 278
- 2017Meta Networkscited by 268
- 2020Self-paced Contrastive Learning with Hybrid Memory for Domain Adaptive Object Re-IDcited by 212
- 2019SimpleShot: Revisiting Nearest-Neighbor Classification for Few-Shot Learningcited by 194
- 2020Cross-Domain Few-Shot Classification via Learned Feature-Wise Transformationcited by 183
- 2018Low-Shot Learning from Imaginary Datacited by 181
- 2018Unsupervised Learning via Meta-Learningcited by 81
- 2018Unsupervised Meta-Learning For Few-Shot Image Classificationcited by 74
- 2019Image Deformation Meta-Networks for One-Shot Learningcited by 56
- 2019Few-Shot Learning via Saliency-guided Hallucination of Samplescited by 49
- 2020DeepEMD: Differentiable Earth Mover's Distance for Few-Shot Learningcited by 49
- 2018Disjoint Label Space Transfer Learning with Common Factorised Spacecited by 13
Built on by
Citing papers per year: 2021: 1; 2022: 0; 2023: 1.
2 papers.