This paper describes a new tech- nique for handling missing multimodal data using a specialized denoising autoencoder: the Multimodal Autoencoder (MMAE).
確定! 回上一頁