[1901.05555] Class-Balanced Loss Based on Effective Number of Samples (arxiv.org) Class-Balanced Loss Based on Effective Number of Samples With the rapid increase of large-scale, real-world datasets, it becomes critical to address the problem of long-tailed data distribution (i.e., a few classes account for most of the data, while most classes are under-represented). Existing solutions typica ar..
아래 자료를 바탕으로 내가 정리한 게시물 https://haystar.tistory.com/79 [논문정리] CoAtNet : Marrying Convolution and Attention for All Data Sizes 논문정보 CoAtNet : Marrying Convolution and Attention for All Data Sizes 논문정리 Abstract 트랜스포머로 인해 컴퓨터 비전에 대한 관심이 높아졌지만, SOTA 컨볼루션망에 비해서는 뒤쳐지고 있다. 이 연구에서는 haystar.tistory.com --------------------------------------------------------------------------------------------------------..