VIME

2000년 도입 · 논문 1편에서 사용

VIME , or Value Imputation and Mask Estimation, is a self- and semi-supervised learning framework for tabular data. It consists of a pretext task of estimating mask vectors from corrupted tabular data in addition to the reconstruction pretext task for self-supervised learning.

출처: VIME: Extending the Success of Self- and Semi-supervised Learning to Tabular Domain

소개 논문: VIME: Extending the Success of Self- and Semi-supervised Learning to Tabular Domain