Learning a Few-shot Embedding Model with Contrastive Learning

02/02/2021

Learning a Few-shot Embedding Model with Contrastive Learning

Chen Liu, Yanwei Fu, Chengming Xu, Siqian Yang, Jilin Li, Chengjie Wang, Li Zhang

Keywords:

Abstract Paper Similar Papers

Abstract: Few-shot learning (FSL) aims to recognize target classes by adapting the prior knowledge learned from source classes. Such knowledge usually resides in a deep embedding model for a general matching purpose of the support and query image pairs. The objective of this paper is to repurpose the contrastive learning for such matching to learn a few-shot embedding model. We make the following contributions: (i) We investigate the contrastive learning with Noise Contrastive Estimation (NCE) in a supervised manner for training a few-shot embedding model; (ii) We propose a novel contrastive training scheme dubbed infoPatch, exploiting the patch-wise relationship to substantially improve the popular infoNCE; (iii) We show that the embedding learned by the proposed infoPatch is more effective; (iv) Our model is thoroughly evaluated on few-shot recognition task; and demonstrates state-of-the-art results on miniImageNet and appealing performance on tieredImageNet, Fewshot-CIFAR100 (FC-100).

The video of this talk cannot be embedded. You can watch it here:

https://slideslive.com/38948106

(Link will open in new window)

0

0

0

0

Share

This is an embedded video. Talk and the respective paper are published at AAAI 2021 virtual conference. If you are one of the authors of the paper and want to manage your upload, see the question "My papertalk has been externally embedded..." in the FAQ section.

Comments

Post Comment

no comments yet

Similar Papers

03/05/2021

Learning and Evaluating Representations for Deep One-Class Classification

Kihyuk Sohn, Chun-Liang Li, Jinsung Yoon and
Minho Jin, Tomas Pfister

Keywords Paper

self-supervised learning, deep one-class classification

0

0

0

1

5:13

14/06/2020

Few-Shot Learning via Embedding Adaptation With Set-to-Set Functions

Han-Jia Ye, Hexiang Hu, De-Chuan Zhan, Fei Sha

Keywords Paper

few-shot learning, meta-learning, embedding learning, embedding adaptation, set-to-set

0

0

0

0

1:04

03/05/2021

$i$-Mix: A Domain-Agnostic Strategy for Contrastive Representation Learning

Kibok Lee, Yian Zhu, Kihyuk Sohn and
Chun-Liang Li, Jinwoo Shin, Honglak Lee

Keywords Paper

self-supervised learning, unsupervised representation learning, data augmentation, MixUp, contrastive representation learning

0

0

0

0

5:04

05/01/2021

Few-Shot Learning via Feature Hallucination With Variational Inference

Qinxuan Luo, Lingfeng Wang, Jingguo Lv and
Shiming Xiang, Chunhong Pan

Keywords Paper

0

0

0

0

4:56

02/02/2021

Harmonized Dense Knowledge Distillation Training for Multi-Exit Architectures

Xinglu Wang, Yingming Li

Keywords Paper

0

0

0

0

15:12

07/09/2020

Meta-RetinaNet for Few-shot Object Detection

Shaoqi Li, Wenfeng Song, Shuai Li and
Aimin Hao, Hong Qin

Keywords Paper

Few shot, object detection, meta-learning, Meta-RetinaNet, Balanced Loss, coefficient vector

0

0

0

0

8:51

03/05/2021

IEPT: Instance-Level and Episode-Level Pretext Tasks for Few-Shot Learning

Manli Zhang, Jianhong Zhang, Zhiwu Lu and
Tao Xiang, Mingyu Ding, Songfang Huang

Keywords Paper

self-supervised learning, few-shot learning, episode-level pretext task

0

0

0

0

5:03

02/02/2021

DeepCollaboration: Collaborative Generative and Discriminative Models for Class Incremental Learning

Bo Cui, Guyue Hu, Shan Yu

Keywords Paper

0

0

0

0

15:13

06/12/2021

Unleashing the Power of Contrastive Self-Supervised Visual Models via Contrast-Regularized Fine-Tuning

Yifan Zhang, Bryan Hooi, Dapeng Hu and
Jian Liang, Jiashi Feng

Keywords Paper

optimization, machine learning, self-supervised learning, vision, contrastive learning, representation learning, transfer learning

0

0

0

0

14:34

05/01/2021

Intra-Class Part Swapping for Fine-Grained Image Classification

Lianbo Zhang, Shaoli Huang, Wei Liu

Keywords Paper

0

0

0

0

4:43

14/06/2020

Semi-Supervised Semantic Segmentation With Cross-Consistency Training

Yassine Ouali, Céline Hudelot, Myriam Tami

Keywords Paper

semantic segmentation, semi-supervised learning, consistency training, semi-supervised semantic segmentation

0

0

0

0

1:01

03/05/2021

CPR: Classifier-Projection Regularization for Continual Learning

Sungmin Cha, Hsiang Hsu, Taebaek Hwang and
Flavio Calmon, Taesup Moon

Keywords Paper

regularization, wide local minima, continual learning

0

0

0

1

5:21

14/06/2020

Instance Credibility Inference for Few-Shot Learning

Yikai Wang, Chengming Xu, Chen Liu and
Li Zhang, Yanwei Fu

Keywords Paper

few-shot learning, incidental parameters, regularization path, semi-supervised learning, self-taught learning

0

0

0

0

1:01

18/07/2021

Toward Understanding the Feature Learning Process of Self-supervised Contrastive Learning

Zixin Wen, Yuanzhi Li

Keywords Paper

Theory, Deep learning Theory

0

0

0

0

5:48

02/02/2021

Task Cooperation for Semi-Supervised Few-Shot Learning

Han-Jia Ye, Xin-Chun Li, De-Chuan Zhan

Keywords Paper

0

0

0

0

16:06

02/02/2021

Attributes-Guided and Pure-Visual Attention Alignment for Few-Shot Recognition

Siteng Huang, Min Zhang, Yachen Kang, Donglin Wang

Keywords Paper

0

0

0

0

17:04

22/11/2021

Few-shot Action Recognition with Prototype-centered Attentive Learning

Xiatian Zhu, Antoine S Toisoul, Juan-Manuel Perez-Rua and
Li Zhang, Brais Martinez, Tao Xiang

Keywords Paper

Few-shot learning, Video recognition, Action classification, Small training data, Model pre-training, Meta-learning, Transformer, Self-attention learning, Cross-attention learning, Prototype learning, Prototype-centered learning, Hybrid-attention learning

0

0

0

0

2:22

14/06/2020

TransMatch: A Transfer-Learning Scheme for Semi-Supervised Few-Shot Learning

Zhongjie Yu, Lin Chen, Zhongwei Cheng, Jiebo Luo

Keywords Paper

few-shot learning, semi-supervised learning, meta-learning

0

0

0

0

1:01

06/12/2020

Restoring Negative Information in Few-Shot Object Detection

Yukuan Yang, Fangyun Wei, Miaojing Shi, Guoqi Li

Keywords Paper

0

0

0

0

3:24

05/01/2021

Class-Wise Metric Scaling for Improved Few-Shot Classification

Ge Liu, Linglan Zhao, Wei Li and
Dashan Guo, Xiangzhong Fang

Keywords Paper

0

0

0

0

5:01

19/04/2021

Keep learning: Self-supervised meta-learning for learning from inference

Akhil Kedia, Sai Chetan Chinthakindi

Keywords Paper

0

0

0

0

11:27

14/06/2020

Adaptive Subspaces for Few-Shot Learning

Christian Simon, Piotr Koniusz, Richard Nock, Mehrtash Harandi

Keywords Paper

subspace, few, shot, meta, learning, classification

0

0

0

0

1:01

19/08/2021

Novelty Detection via Contrastive Learning with Negative Data Augmentation

Chengwei Chen, Yuan Xie, Shaohui Lin and
Ruizhi Qiao, Jian Zhou, Xin Tan, Yi Zhang, Lizhuang Ma

Keywords Paper

Computer Vision, 2D and 3D Computer Vision, Anomaly/Outlier Detection

0

0

0

0

14:37

03/05/2021

Exploring Balanced Feature Spaces for Representation Learning

Bingyi Kang, Yu Li, Sain Xie and
Zehuan Yuan, Jiashi Feng

Keywords Paper

Representation Learning, Contrastive Learning, Long-Tailed Recognition

0

0

0

0

7:18

03/05/2021

On Data-Augmentation and Consistency-Based Semi-Supervised Learning

Atin Ghosh, alexandre thiery

Keywords Paper

Semi-Supervised Learning, Regularization, Data augmentation

0

0

0

0

4:42

18/07/2021

Improved OOD Generalization via Adversarial Training and Pretraing

Mingyang Yi, Lu Hou, Jiacheng Sun and
Lifeng Shang, Xin Jiang, Qun Liu, Zhiming Ma

Keywords Paper

Theory, Deep learning Theory

0

0

0

0

4:11

12/07/2020

Feature-map-level Online Adversarial Knowledge Distillation

Inseop Chung, SeongUk Park, Kim Jangho, NOJUN KWAK

Keywords Paper

Applications - Computer Vision

0

0

0

0

14:06

26/04/2020

Frequency-based Search-control in Dyna

Yangchen Pan, Jincheng Mei, Amir-massoud Farahmand

Keywords Paper

Model-based reinforcement learning, search-control, Dyna, frequency of a signal

0

0

0

0

4:32

02/02/2021

A Simple and Effective Self-Supervised Contrastive Learning Framework for Aspect Detection

Tian Shi, Liuqing Li, Ping Wang, Chandan K. Reddy

Keywords Paper

0

0

0

0

19:21

14/06/2020

Better Captioning With Sequence-Level Exploration

Jia Chen, Qin Jin

Keywords Paper

caption, sequece-level, diversity, precision

0

0

0

0

0:57

14/06/2020

CONSAC: Robust Multi-Model Fitting by Conditional Sample Consensus

Florian Kluger, Eric Brachmann, Hanno Ackermann and
Carsten Rother, Michael Ying Yang, Bodo Rosenhahn

Keywords Paper

robust estimator, reinforcement learning, self-supervised, unsupervised, multi-model, ransac, dataset, vanishing points, homography, 3d reconstruction

0

0

0

0

1:00

26/04/2020

Progressive learning and disentanglement of hierarchical representations

Zhiyuan Li, Jaideep Vitthal Murkute, Prashnna Kumar Gyawali, Linwei Wang

Keywords Paper

generative model, disentanglement, progressive learning, VAE

0

0

0

0

5:06

05/01/2021

Towards Fair Cross-Domain Adaptation via Generative Learning

Tongxin Wang, Zhengming Ding, Wei Shao and
Haixu Tang, Kun Huang

Keywords Paper

0

0

0

0

4:56

18/07/2021

Function Contrastive Learning of Transferable Meta-Representations

Waleed Gondal, Shruti Joshi, Nasim Rahaman and
Stefan Bauer, Manuel Wuthrich, Bernhard Schölkopf

Keywords Paper

Algorithms, Multitask, Transfer, and Meta Learning

0

0

0

0

5:46

06/12/2021

$(\textrm{Implicit})^2$: Implicit Layers for Implicit Representations

Zhichun Huang, Shaojie Bai, J. Zico Kolter

Keywords Paper

deep learning, representation learning

1

0

0

1

12:23

06/12/2021

Adversarial Reweighting for Partial Domain Adaptation

Xiang Gu, Xi Yu, yan yang and
Jian Sun, Zongben Xu

Keywords Paper

domain adaptation

0

0

0

1

9:03

26/04/2020

Meta-Q-Learning

Rasool Fakoor, Pratik Chaudhari, Stefano Soatto, Alexander J. Smola

Keywords Paper

meta reinforcement learning, propensity estimation, off-policy

0

0

0

0

15:50

08/12/2020

Mixup-Transformer: Dynamic Data Augmentation for NLP Tasks

Lichao Sun, Congying Xia, Wenpeng Yin and
Tingting Liang, Philip Yu, Lifang He

Keywords Paper

0

0

0

0

9:52

03/05/2021

Dataset Meta-Learning from Kernel Ridge-Regression

Timothy Nguyen, Zhourong Chen, Jaehoon Lee

Keywords Paper

dataset corruption, infinite-width networks, neural kernels, kernel-ridge regression, dataset compression, dataset distillation, meta-learning

0

0

0

0

4:59

03/05/2021

Adaptive and Generative Zero-Shot Learning

Yu-Ying Chou, Hsuan-Tien (Tien) Lin, Tyng-Luh Liu

Keywords Paper

Generalized zero-shot learning, mixup

0

0

0

0

5:18