39 citations · 119 across the 12 of their papers we have counts for
6 papers · 1 filter
Does Deep Learning Learn to Abstract? A Systematic Probing Framework
Shengnan An, Zeqi Lin, Bei Chen +3
Abstraction is a desirable capability for deep learning models, which means to induce abstract concepts from concrete instances and flexibly apply them beyond the learning context.…
Make Heterophily Graphs Better Fit GNN: A Graph Rewiring Approach
Wendong Bi, Lun Du, Qiang Fu +3
Graph Neural Networks (GNNs) are popular machine learning methods for modeling graph data. A lot of GNNs perform well on homophily graphs while having unsatisfactory performance on…
Learning Rate Perturbation: A Generic Plugin of Learning Rate Schedule towards Flatter Local Minima
Hengyu Liu, Qiang Fu, Lun Du +4
Learning rate is one of the most important hyper-parameters that has a significant influence on neural network training. Learning rate schedules are widely used in real practice to…
Neuron with Steady Response Leads to Better Generalization
Qiang Fu, Lun Du, Haitao Mao +4
Regularization can mitigate the generalization gap between training and inference by introducing inductive bias. Existing works have already proposed various inductive biases from…
GBK-GNN: Gated Bi-Kernel Graph Neural Networks for Modeling Both Homophily and Heterophily
Lun Du, Xiaozhou Shi, Qiang Fu +4
Graph Neural Networks (GNNs) are widely used on a variety of graph-based machine learning tasks. For node-level tasks, GNNs have strong power to model the homophily property of gra…
Neuron Campaign for Initialization Guided by Information Bottleneck Theory
Haitao Mao, Xu Chen, Qiang Fu +3
Initialization plays a critical role in the training of deep neural networks (DNN). Existing initialization strategies mainly focus on stabilizing the training process to mitigate…