Graph Laplacian interpolating activation replaces softmax in DNNs and improves natural accuracy, robust accuracy, and data efficiency.
Deep Residual Learning and PDEs on Manifold
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
In this paper, we formulate the deep residual network (ResNet) as a control problem of transport equation. In ResNet, the transport equation is solved along the characteristics. Based on this observation, deep neural network is closely related to the control problem of PDEs on manifold. We propose several models based on transport equation and Hamilton-Jacobi equation. The discretization of these PDEs on point cloud is also discussed.
fields
cs.LG 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Graph Interpolating Activation Improves Both Natural and Robust Accuracies in Data-Efficient Deep Learning
Graph Laplacian interpolating activation replaces softmax in DNNs and improves natural accuracy, robust accuracy, and data efficiency.