Piecewise linear activations substantially shape the loss surfaces of neural networks

Abstract

Understanding the loss surface of a neural network is fundamentally important to the understanding of deep learning. This paper presents how piecewise linear activation functions substantially shape the loss surfaces of neural networks. We first prove that {\it the loss surfaces of many neural networks have infinite spurious local minima} which are defined as the local minima with higher empirical risks than the global minima. Our result demonstrates that the networks with piecewise linear activations possess substantial differences to the well-studied linear neural networks. This result holds for any neural network with arbitrary depth and arbitrary piecewise linear activation functions (excluding linear functions) under most loss functions in practice. Essentially, the underlying assumptions are consistent with most practical circumstances where the output layer is narrower than any hidden layer. In addition, the loss surface of a neural network with piecewise linear activations is partitioned into multiple smooth and multilinear cells by nondifferentiable boundaries. The constructed spurious local minima are concentrated in one cell as a valley: they are connected with each other by a continuous path, on which empirical risk is invariant. Further for one-hidden-layer networks, we prove that all local minima in a cell constitute an equivalence class; they are concentrated in a valley; and they are all global minima in the cell.

arXiv commentPublished as a conference paper at ICLR 2020

Cited by in corpus (10)

• Recent advances in deep learning theory
• Understanding Deep Learning via Decision Boundary
• Truth or Backpropaganda? An Empirical Investigation of Deep Learning Theory
• Tighter Generalization Bounds for Iterative Differentially Private Learning Algorithms
• Training Two-Layer ReLU Networks with Gradient Descent is Inconsistent
• Sub-Optimal Local Minima Exist for Neural Networks with Almost All Non-Linear Activations
• Analytic function approximation by path norm regularized deep networks
• Neural networks behave as hash encoders: An empirical study
• When Are Solutions Connected in Deep Networks?
• Spurious Local Minima Are Common for Deep Neural Networks with Piecewise Linear Activations