4 papers
Sphere Retraction Normalizations
Jie Zhang, Cheng-Fang Su, Yi-Jui Huang +1
Residual connections are the de facto mechanism for training deep neural networks stably. Geodesic Normalization (GeoNorm) recasts them on a Riemannian manifold, orthogonalizing ea…
Towards Efficient Constraint Handling in Neural Solvers for Routing Problems
Jieyi Bi, Zhiguang Cao, Jianan Zhou +5
Neural solvers have achieved impressive progress in addressing simple routing problems, particularly excelling in computational efficiency. However, their advantages under complex…
Converting Transformers into DGNNs Form
Jie Zhang, Mao-Hsuan Mao, Bo-Wei Chiu +1
Recent advances in deep learning have established Transformer architectures as the predominant modeling paradigm. Central to the success of Transformers is the self-attention mecha…
From ChebNet to ChebGibbsNet
Jie Zhang, Min-Te Sun
Recent advancements in Spectral Graph Convolutional Networks (SpecGCNs) have led to state-of-the-art performance in various graph representation learning tasks. To exploit the pote…