Neural Normalized Cut: A differential and generalizable approach for spectral clustering

Wei He, Shangzhi Zhang, Chun-Guang Li, Xianbiao Qi, Rong Xiao, Jun Guo

Abstract: Highlights•Relax NCut into a continuous problem with orthogonality penalty and box constraints.•Use neural network to reparameterize the clustering membership and train it with SGD.•Conduct extensive experiments to show the effectiveness and the generalization ability.