Self-supervised Learning: Generative or Contrastive

Liu, Xiao; Zhang, Fanjin; Hou, Zhenyu; Li, Mian; Wang, Zhaoyu; Zhang, Jing; Tang, Jie

doi:10.1109/tkde.2021.3090866

Cited by 817 publications

(513 citation statements)

References 62 publications

Supporting

Mentioning

405

Contrasting

Unclassified

Order By: Relevance

“…Moreover, fine-tuning with labeled samples can follow a self-supervised representation \metric learning procedure, where positive and negative pairs are generated automatically from a large amount of unlabeled images and are used to train a deep learning backbone (e.g., ResNet) to minimize the feature distance of positive pair samples and maximize the feature distance for negative samples [153,154]. Researchers [155] proposed a metric learning model named discriminative CNN(D-CNN) to perform metric learning and obtained state-of-the-art performance on three public RS datasets for scene classification tasks, showing promising results of using self-supervised learning approaches for landcover mapping task.…”

Section: Model Fine-tuningmentioning

confidence: 99%

A Review of Landcover Classification with Very-High Resolution Remotely Sensed Optical Images—Analysis Unit, Model Scalability and Transferability

Qin

Liu

2022

Remote Sensing

View full text Add to dashboard Cite

As an important application in remote sensing, landcover classification remains one of the most challenging tasks in very-high-resolution (VHR) image analysis. As the rapidly increasing number of Deep Learning (DL) based landcover methods and training strategies are claimed to be the state-of-the-art, the already fragmented technical landscape of landcover mapping methods has been further complicated. Although there exists a plethora of literature review work attempting to guide researchers in making an informed choice of landcover mapping methods, the articles either focus on the review of applications in a specific area or revolve around general deep learning models, which lack a systematic view of the ever advancing landcover mapping methods. In addition, issues related to training samples and model transferability have become more critical than ever in an era dominated by data-driven approaches, but these issues were addressed to a lesser extent in previous review articles regarding remote sensing classification. Therefore, in this paper, we present a systematic overview of existing methods by starting from learning methods and varying basic analysis units for landcover mapping tasks, to challenges and solutions on three aspects of scalability and transferability with a remote sensing classification focus including (1) sparsity and imbalance of data; (2) domain gaps across different geographical regions; and (3) multi-source and multi-view fusion. We discuss in detail each of these categorical methods and draw concluding remarks in these developments and recommend potential directions for the continued endeavor.

show abstract

Section: Model Fine-tuningmentioning

confidence: 99%

A Review of Landcover Classification with Very-High Resolution Remotely Sensed Optical Images—Analysis Unit, Model Scalability and Transferability

Qin

Liu

2022

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…There are two approaches: one that uses statistical learning, developed since the beginning of AI in the 1950's, and the other exploiting logical learning where rules are defined to create a description of the element of interest. It is assumed that input data are from outcomes or from rules, and it is always provided by humans [30,34,35].…”

Section: Epidemiological Data Analysismentioning

confidence: 99%

AI & COVID-19

Bacciu

Girardi²,

Maratea

et al. 2022

View full text Add to dashboard Cite

The COVID-19 pandemic has influenced our lives significantly since March 2020, and a number of initiatives have been put forward in order to tackle its effects, including those focused on technological solutions. In this paper, we present one of such initiatives, i.e. the CLAIRE’s taskforce on AI and COVID-19, in which Artificial Intelligence methodologies and tools are being developed to help the society contrasting the pandemic. We present the different lines of development within the taskforce, some fields in which they are used, and draw few recommendations.

show abstract

“…In contrast, as a new paradigm between unsupervised and supervised learning, SSL can generate labels based on the property of unlabeled data itself to train the neural network in a supervised manner similar to natural learning experiences. With excellent performance on representation learning and dealing with the issue of unlabelled data, SSL [20][21][22] has been successfully implemented in a wide range of fields, including image recognition 23 , audio representation 24 , computer vision 25 , document reconstruction 26 , atmosphere 27 , astronomy 28 , medical 29 , person re-identification 30 , remote sensing 31 , robotics 32 , omnidirectional imaging 33 , manufacturing 34 , nano-photonics 35 , and civil engineering 36 , etc. However, this method has not been formally attempted in material science.…”

Section: High-efficient Low-cost Characterization Of Materials Proper...mentioning

confidence: 99%

High-efficient low-cost characterization of materials properties using domain-knowledge-guided self-supervised learning

Xie

Yao

Mao

et al. 2022

Preprint

View full text Add to dashboard Cite

Modern AI-assisted approaches have helped material scientists revolutionize their abilities to better understand the properties of materials. However, current machine learning (ML) models would perform awful for materials with a lengthy production window and a complex testing procedure because only a limited amount of data can be produced to feed the model. Here, we introduce self-supervised learning (SSL) to address the issue of lacking labeled data in material characterization. We propose a generalized SSL-based framework with domain knowledge and demonstrate its robustness to predict the properties of a candidate material with the fewest data. Our numerical results show that the performance of the proposed SSL model can match the commonly-used supervised learning (SL) model with only 5 % of data, and the SSL model is also proven with ease of implementation. Our study paves the way to expand further the usability of ML tools for a broader material science community.

show abstract

Self-supervised Learning: Generative or Contrastive

Cited by 817 publications

References 62 publications

A Review of Landcover Classification with Very-High Resolution Remotely Sensed Optical Images—Analysis Unit, Model Scalability and Transferability

A Review of Landcover Classification with Very-High Resolution Remotely Sensed Optical Images—Analysis Unit, Model Scalability and Transferability

AI & COVID-19

High-efficient low-cost characterization of materials properties using domain-knowledge-guided self-supervised learning

Contact Info

Product

Resources

About