본문으로 바로가기

[개념정리] Pretext task 란?

category AI/CV & SSL 2019. 10. 31. 16:52
336x280(권장), 300x250(권장), 250x250, 200x200 크기의 광고 코드만 넣을 수 있습니다.

 

 

 

 

 

A pretext task is used in self-supervised learning to generate useful feature representations, where “useful” means a representation that should be easily adaptable for other tasks, unknown during training time.

 

Pretext Task: Pretext tasks are pre-designed tasks for networks to solve, and visual features are learned by learning objective functions of pretext tasks.

 

 

 

https://stats.stackexchange.com/questions/404602/pretext-task-in-computer-vision

 

Pretext Task in Computer Vision

I am new to Computer Vision. I am reading many papers and i see the term "pretext task". Can anyone explain what exactly it means. Thanks in Advance.

stats.stackexchange.com

 

 

 

 

 

'AI > CV & SSL' 카테고리의 다른 글

[개념정리] downstream task란?  (0) 2019.11.07
[개념정리] co-adaptation이란?  (0) 2019.11.01
[개념정리] Ablation Study 란?  (0) 2019.10.31
[개념정리] Generalization & Optimization  (0) 2019.10.30
[개념정리] 과적합이란?  (0) 2019.09.20