Timezone: »

 
Oral
MSplit LBI: Realizing Feature Selection and Dense Estimation Simultaneously in Few-shot and Zero-shot Learning
Bo Zhao · Xinwei Sun · Yanwei Fu · Yuan Yao · Yizhou Wang

Wed Jul 11 02:30 AM -- 02:40 AM (PDT) @ K1 + K2
in Feature Selection 1 »
It is one typical and general topic of learning a good embedding model to efficiently learn the representation coefficients between two spaces/subspaces. To solve this task, $L_{1}$ regularization is widely used for the pursuit of feature selection and avoiding overfitting, and yet the sparse estimation of features in $L_{1}$ regularization may cause the underfitting of training data. $L_{2}$ regularization is also frequently used, but it is a biased estimator. In this paper, we propose the idea that the features consist of three orthogonal parts, \emph{namely} sparse strong signals, dense weak signals and random noise, in which both strong and weak signals contribute to the fitting of data. To facilitate such novel decomposition, \emph{MSplit} LBI is for the first time proposed to realize feature selection and dense estimation simultaneously. We provide theoretical and simulational verification that our method exceeds $L_{1}$ and $L_{2}$ regularization, and extensive experimental results show that our method achieves state-of-the-art performance in the few-shot and zero-shot learning.
[ PDF]  [ Video] 

Author Information

Bo Zhao (Peking University)
Xinwei Sun (Peking University)
Yanwei Fu (Fudan university)
Yuan Yao (Hong Kong Science Tech)
Yizhou Wang (Peking University)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors