Few-Shot Learning

A curated list of resources including papers, comparitive results on standard datasets and relevant links pertaining to few-shot learning.

Contributing

Contributions are welcome. If you have suggestions for new sections or valuable works to be included, please feel free to raise an issue and discuss in issue module.

Table of Contents

• Papers
• Datasets
• Starter code for FSL
• Other Resources

Few-Shot Learning (Classification)

Papers

[1] Multi-domain based FSL [Website]

• Meta-Dataset: Eleni Triantafillou, Tyler Zhu, Vincent Dumoulin, Pascal Lamblin, Utku Evci, Kelvin Xu, Ross Goroshin, Carles Gelada, Kevin Swersky, Pierre-Antoine Manzagol, and Hugo Larochelle. "Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples." ICLR (2020). [pdf] [code].

• SimpleCNAPS: Peyman Bateni, Raghav Goyal, Vaden Masrani, Frank Wood, Leonid Sigal. "Improved Few-Shot Visual Classification." CVPR (2020). [pdf] [code].

• SUR: Nikita Dvornik, Cordelia Schmid, and Julien Mairal. "Selecting Relevant Features from a Multi-domain Representation for Few-shot Classification." ECCV (2020). [pdf] [code].

• URT: Lu Liu, William L. Hamilton, Guodong Long, Jing Jiang, and Hugo Larochelle. "A Universal Representation Transformer Layer for Few-Shot Image Classification." ICLR(2021). [pdf][code].

• FLUTE: Eleni Triantafillou, Hugo Larochelle, Richard Zemel, Vincent Dumoulin. "Learning a Universal Template for Few-shot Dataset Generalization." ICML(2021). [pdf].

• URL: Wei-Hong Li, Xialei Liu, Hakan Bilen. "Universal Representation Learning from Multiple Domains for Few-shot Classification." ICCV(2021). [pdf].

• ITA: Wei-Hong Li, Xialei Liu, Hakan Bilen. "Improving Task Adaptation for Cross-domain Few-shot Learning." arXiv(2021). [pdf].

[2] Singal-domain based FSL

• LibFewShot: wenbin Li, Chuanqi Dong, Pinzhuo Tian, Tiexin Qin, Xuesong Yang, Ziyi Wang, Jing Huo, Yinghuan Shi, Lei Wang, Yang Gao, Jiebo Luo. "LibFewShot: A Comprehensive Library for Few-shot Learning." arXiv(2021). [pdf] [code]. <an emperical survey>

  

Meta-Learning + data augmentation

• KIP: Timothy Nguyen, Zhourong Chen, and Jaehoon Lee. "Dataset Meta-Learning from Kernel Ridge-Regression." ICLR(2021). [pdf] [code].

  

• ICLR withdraw: Jialin Liu, Fei Chao, and Chih-Min Lin. "Task Level Data Augmentation for Meta-Learning." ICLR withdraw(2021). [pdf].

• ICLR withdraw: Renkun Ni, Micah Goldblum, Amr Sharaf, Kezhi Kong, and Tom Goldstein. "Data Augmentation for Meta-Learning." ICLR withdraw(2021). [pdf].

• ICLR reject: Xingjian Li, Haoyi Xiong, Haozhe An, Cheng-zhong Xu, and Dejing Dou. "XMixup: Efficient Transfer Learning with Auxiliary Samples by Cross-Domain Mixup." ICLR reject(2021). [pdf].

• ICLR reject: Georgios Batzolis, Alberto Bernacchia, Da-shan Shiu, Michael Bromberg, and Alexandru Cioba. "Optimal allocation of data across training tasks in meta-learning." ICLR reject(2021). [pdf].

ICLR 2021

• DC: Shuo Yang, Lu Liu, and Min Xu. "Free Lunch for Few-shot Learning: Distribution Calibration." ICLR oral(2021). [pdf] [code].

  

• STARTUP: Cheng Perng Phoo, and Bharath Hariharan. "Self-training For Few-shot Transfer Across Extreme Task Differences." ICLR oral(2021). [pdf].

  

• CPM: Mengye Ren, Michael Louis Iuzzolino, Michael Curtis Mozer, and Richard Zemel. "Wandering within a world: Online contextualized few-shot learning." ICLR(2021). [pdf].

  

• THEORY: Simon Shaolei Du, Wei Hu, Sham M. Kakade, Jason D. Lee, and Qi Lei. "Few-Shot Learning via Learning the Representation, Provably." ICLR(2021). [pdf].

• URT: Lu Liu, William L. Hamilton, Guodong Long, Jing Jiang, and Hugo Larochelle. "A Universal Representation Transformer Layer for Few-Shot Image Classification." ICLR(2021). [pdf].

  

• COMET: Kaidi Cao, Maria Brbic, and Jure Leskovec. "Concept Learners for Few-Shot Learning." ICLR(2021). [pdf].

  

• IEPT: Manli Zhang, Jianhong Zhang, Zhiwu Lu, Tao Xiang, Mingyu Ding, and Songfang Huang. "IEPT: Instance-Level and Episode-Level Pretext Tasks for Few-Shot Learning." ICLR(2021). [pdf].

  

• IDLVQ-C: Kuilin Chen, and Chi-Guhn Lee. "Incremental few-shot learning via vector quantization in deep embedded space." ICLR(2021). [pdf]. <LVQ>

• SLE: Bingchen Liu, Yizhe Zhu, Kunpeng Song, and Ahmed Elgammal. "Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis." ICLR(2021). [pdf].

• repurposing MAML: Namyeong Kwon, Hwidong Na, Gabriel Huang, and Simon Lacoste-Julien. "Repurposing Pretrained Models for Robust Out-of-domain Few-Shot Learning." ICLR(2021). [pdf] [code].

  

• MELR: Nanyi Fei, Zhiwu Lu, Tao Xiang, and Songfang Huang. "MELR: Meta-Learning via Modeling Episode-Level Relationships for Few-Shot Learning." ICLR(2021). [pdf].

  

• MB(Supervised): Mihir Prabhudesai, Shamit Lal, Darshan Patil, Hsiao-Yu Tung, Adam W Harley, and Katerina Fragkiadaki. "Disentangling 3D Prototypical Networks for Few-Shot Concept Learning." ICLR(2021). [pdf].

• MetaNorm: Yingjun Du, Xiantong Zhen, Ling Shao, and Cees G. M. Snoek. "MetaNorm: Learning to Normalize Few-Shot Batches Across Domains." ICLR(2021). [pdf].<Meta-Dataset>

  

• ConstellationNet: Weijian Xu, yifan xu, Huaijin Wang, and Zhuowen Tu. "Constellation Nets for Few-Shot Learning." ICLR(2021). [pdf].

  

• OVE PG GP + Cosine: Jake Snell, and Richard Zemel. "Bayesian Few-Shot Classification with One-vs-Each Pólya-Gamma Augmented Gaussian Processes." ICLR(2021). [pdf].

• BOIL: Jaehoon Oh, Hyungjun Yoo, ChangHwan Kim, and Se-Young Yun. "BOIL: Towards Representation Change for Few-shot Learning." ICLR(2021). [pdf].

  

• FBNet: Zhipeng Bao, Yu-Xiong Wang, and Martial Hebert. "Bowtie Networks: Generative Modeling for Joint Few-Shot Recognition and Novel-View Synthesis." ICLR(2021). [pdf]. <uncertainty quantification>

  

NIPS 2020

• DKT: Patacchiola, Massimiliano and Turner, Jack and Crowley, Elliot J and O'Boyle, Michael and Storkey, Amos. "Bayesian Meta-Learning for the Few-Shot Setting via Deep Kernels." NIPS (2020). [pdf] [code].

ECCV 2020

• MABAS: Jaekyeom Kim, Hyoungseok Kim, and Gunhee Kim. "Model-Agnostic Boundary-Adversarial Sampling for Test-Time Generalization in Few-Shot learning." ECCV (2020). [pdf] [code].

  

• centroid alignment: Arman Afrasiyabi, Jean-François Lalonde, and Christian Gagné. "Associative Alignment for Few-shot Image Classification." ECCV (2020). [pdf] [code].

  

• TAFSSL: Moshe Lichtenstein, Prasanna Sattigeri, Rogerio Feris, Raja Giryes, and Leonid Karlinsky. "TAFSSL: Task-Adaptive Feature Sub-Space Learning for few-shot classification." ECCV (2020). [pdf]. <Semi-supervised><transductive>

  

• BD-CSPN: Jinlu Liu, Liang Song, and Yongqiang Qin. "Prototype Rectification for Few-Shot Learning." ECCV (2020). [pdf]. <Meta-Dataset><transductive>

  

• SSL-FSL: Jong-Chyi Su, Subhransu Maji, and Bharath Hariharan. "When Does Self-supervision Improve Few-shot Learning?." ECCV (2020). [pdf]. <self-supervised>

  

• IDA: Qing Liu, Orchid Majumder, Alessandro Achille, Avinash Ravichandran, Rahul Bhotika, and Stefano Soatto. "Incremental Few-Shot Meta-Learning via Indirect Discriminant Alignment." ECCV (2020). [pdf]. <Incremental>

  

• LS-FSL: Shuo Wang, Jun Yue, Jianzhuang Liu, Qi Tian, and Meng Wang. "Large-Scale Few-Shot Learning via Multi-Modal Knowledge Discovery." ECCV (2020). [pdf].

  

• SUR: Nikita Dvornik, Cordelia Schmid, and Julien Mairal. "Selecting Relevant Features from a Multi-domain Representation for Few-shot Classification." ECCV (2020). [pdf] [code]. <Meta-Dataset>

  

• LR-distill: Yonglong Tian, Yue Wang, Dilip Krishnan, Joshua B. Tenenbaum, and Phillip Isola. "Rethinking Few-shot Image Classification: A Good Embedding is All You Need?." ECCV (2020). [pdf] [code]. <self-supervised> <Meta-Dataset>

  

• E³BM: Yaoyao Liu, Bernt Schiele, and Qianru Sun. "An Ensemble of Epoch-wise Empirical Bayes for Few-shot Learning." ECCV (2020). [pdf] [code]. <transductive and inductive>

  

• Data Design: Othman Sbai, Camille Couprie, and Mathieu Aubry. "Impact of base dataset design on few-shot image classification." ECCV (2020). [pdf] [code]. <base training set selection>

  

• SEN: Van Nhan Nguyen, Sigurd Løkse, Kristoffer Wickstrøm, Michael Kampffmeyer, Davide Roverso, and Robert Jenssen. "SEN: A Novel Feature Normalization Dissimilarity Measure for Prototypical Few-Shot Learning Networks." ECCV (2020). [pdf].

  

• EPNet: Pau Rodríguez, Issam Laradji, Alexandre Drouin, and Alexandre Lacoste. "Embedding Propagation: Smoother Manifold for Few-Shot Classification." ECCV (2020). [pdf] [code]. <Semi-supervised><transductive>

  

• BSCD-FSL: Yunhui Guo, Noel C. Codella, Leonid Karlinsky, and James V. Codella, John R. Smith, Kate Saenko, Tajana Rosing, Rogerio Feris. "A Broader Study of Cross-Domain Few-Shot Learning." ECCV (2020). [pdf] [code]. <transductive and inductive>

  

• DeepCaps-FSL: Fangyu Wu, Jeremy S.Smith, Wenjin Lu, Chaoyi Pang, and Bailing Zhang. "Attentive Prototype Few-shot Learning with Capsule Network-based Embedding." ECCV (2020). [pdf]. <transductive>

  

• Neg-Cosine: Bin Liu, Yue Cao, Yutong Lin, Qi Li, Zheng Zhang, Mingsheng Long, and Han Hu. "Negative Margin Matters: Understanding Margin in Few-shot Classification." ECCV (2020). [pdf] [code].

  

ICLR 2020

• ProtoNet-EST: Tianshi Cao, Marc T Law, and Sanja Fidler. "A Theoretical Analysis of the Number of Shots in Few-Shot Learning." ICLR (2020). [pdf].

• Transductive fine-tuning: Guneet Singh Dhillon, Pratik Chaudhari, Avinash Ravichandran, and Stefano Soatto. "A Baseline for Few-Shot Image Classification." ICLR (2020). [pdf]. <transductive>

• LFT-FSL: Hung-Yu Tseng, Hsin-Ying Lee, Jia-Bin Huang, and Ming-Hsuan Yang. "Cross-Domain Few-Shot Classification via Learned Feature-Wise Transformation." ICLR (2020). [pdf] [code].

  

• Meta-Dataset: Eleni Triantafillou, Tyler Zhu, Vincent Dumoulin, Pascal Lamblin, Utku Evci, Kelvin Xu, Ross Goroshin, Carles Gelada, Kevin Swersky, Pierre-Antoine Manzagol, and Hugo Larochelle. "Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples." ICLR (2020). [pdf] [code].

• SIB: Shell Xu Hu, Pablo Garcia Moreno, Yang Xiao, Xi Shen, Guillaume Obozinski, Neil Lawrence, and Andreas Damianou. "Empirical Bayes Transductive Meta-Learning with Synthetic Gradients." ICLR (2020). [pdf] [code]. <transductive>

  

CVPR 2020

• Selection-FSL: Linjun Zhou, Peng Cui, Xu Jia, Shiqiang Yang, and Qi Tian. "Learning to Select Base Classes for Few-Shot Classification." CVPR (2020). [pdf]. <selection>

• DSN: Christian Simon, Piotr Koniusz, Richard Nock, and Mehrtash Harandi. "Adaptive Subspaces for Few-Shot Learning." CVPR (2020). [pdf] [code]. <Semi-supervised><supervised>

  

• Few-Shot Open-set: Bo Liu, Hao Kang, Haoxiang Li, Gang Hua, and Nuno Vasconcelos. "Few-Shot Open-Set Recognition using Meta-Learning." CVPR (2020). [pdf]. <Weakly Supervised>

  

• FEAT: Han-Jia Ye, Hexiang Hu, De-Chuan Zhan, and Fei Sha. "Few-Shot Learning via Embedding Adaptation with Set-to-Set Functions." CVPR (2020). [pdf] [code]. <Transformer> <inductive+transductive>

  

• Simple CNAPS: Bateni, Peyman and Goyal, Raghav and Masrani, Vaden and Wood, Frank and Sigal, Leonid. "Improved Few-Shot Visual Classification." CVPR (2020). [pdf] <Meta-Dataset>

ICML 2020

• TaskNorm: Bronskill, John and Gordon, Jonathan and Requeima, James and Nowozin, Sebastian and Turner, Richard. "TaskNorm: Rethinking Batch Normalization for Meta-Learning." ICML (2020). [pdf] [code].<Meta-Dataset>

arxiv 2020

• Meta-Baseline: Yinbo Chen, Xiaolong Wang, Zhuang Liu, Huijuan Xu, and Trevor Darrell. "A New Meta-Baseline for Few-Shot Learning" arxiv (2020). [pdf] [code].<Meta-Dataset>

• HPO: Saikia, Tonmoy and Brox, Thomas and Schmid, Cordelia. "Optimized Generic Feature Learning for Few-shot Classification across Domains." arxiv (2020). [pdf].<Meta-Dataset>

NIPS 2019

• CNAPS: Jaekyeom Kim, Hyoungseok Kim, and Gunhee Kim. "Fast and Flexible Multi-Task Classification using Conditional Neural Adaptive Processes." NIPS (2019). [pdf] [code].<Meta-Dataset>

  

• LST: Li, Xinzhe and Sun, Qianru and Liu, Yaoyao and Zhou, Qin and Zheng, Shibao and Chua, Tat-Seng and Schiele, Bernt. "Learning to self-train for semi-supervised few-shot classification." NIPS (2019). [pdf] [code].

CVPR 2019

Li, Aoxue and Luo, Tiange and Lu, Zhiwu and Xiang, Tao and Wang, Liwei. "Large-scale few-shot learning: Knowledge transfer with class hierarchy." CVPR(2019). [pdf] [code].

Datasets

• miniImageNet: [link]
• tieredImageNet: [link]
• CUB: [link]
• Meta-Dataset: [link]
• CIFAR-FS: [link]
• FC100: [link]

There are two datasets, Categories: 17 and 102. [link]

Starter Code

There are several backbones:

• Conv4: [link]
• Resnet10: [link]
• Resnet18: [link]
• WRN: [link]

Other resources

Please click [here] for few-shot classification leaderboard.

License