Improving Dropout in Graph Convolutional Networks for Recommendation via Contrastive Loss.

Hiroki Okamura, Keisuke Maeda, Ren Togo, Takahiro Ogawa, Miki Haseyama

Browse the full ICASSP paper archive.