Deep Neural Nets With Interpolating Function As Output Activation

Authors:
Bao Wang UCLA
Xiyang Luo Google
Zhen Li Hong Kong University of Science & Technology
Wei Zhu Duke University
Zuoqiang Shi zqshi@mail.tsinghua.edu.cn
Stanley Osher UCLA

Introduction:

The authors replace the output layer of deep neural nets, typically the softmax function, by a novel interpolating function.And the authors propose end-to-end training and testing algorithms for this new architecture.

Abstract:

We replace the output layer of deep neural nets, typically the softmax function, by a novel interpolating function. And we propose end-to-end training and testing algorithms for this new architecture. Compared to classical neural nets with softmax function as output activation, the surrogate with interpolating function as output activation combines advantages of both deep and manifold learning. The new framework demonstrates the following major advantages: First, it is better applicable to the case with insufficient training data. Second, it significantly improves the generalization accuracy on a wide variety of networks. The algorithm is implemented in PyTorch, and the code is available at https://github.com/BaoWangMath/DNN-DataDependentActivation.

You may want to know: