Landscape Connectivity and Dropout Stability of SGD Solutions for Over-parameterized Neural Networks
arXiv:1912.10095
Abstract
The optimization of multilayer neural networks typically leads to a solution with zero training error, yet the landscape can exhibit spurious local minima and the minima can be disconnected. In this paper, we shed light on this phenomenon: we show that the combination of stochastic gradient descent (SGD) and over-parameterization makes the landscape of multilayer neural networks approximately connected and thus more favorable to optimization. More specifically, we prove that SGD solutions are connected via a piecewise linear path, and the increase in loss along this path vanishes as the number of neurons grows large. This result is a consequence of the fact that the parameters found by SGD are increasingly dropout stable as the network becomes wider. We show that, if we remove part of the neurons (and suitably rescale the remaining ones), the change in loss is independent of the total number of neurons, and it depends only on how many neurons are left. Our results exhibit a mild dependence on the input dimension: they are dimension-free for two-layer networks and depend linearly on the dimension for multilayer networks. We validate our theoretical findings with numerical experiments for different architectures and classification tasks.
arXiv commentProceedings of the 37th International Conference on Machine Learning (ICML)
References in corpus (25)
- Rethinking the Value of Network Pruning
- The Loss Surfaces of Multilayer Networks
- A Convergence Theory for Deep Learning via Over-Parameterization
- A Mean Field View of the Landscape of Two-Layers Neural Networks
- Identifying and attacking the saddle point problem in high-dimensional non-convex optimization
- The State of Sparsity in Deep Neural Networks
- Variational Dropout Sparsifies Deep Neural Networks
- Learning Overparameterized Neural Networks via Stochastic Gradient Descent on Structured Data
- On Lazy Training in Differentiable Programming
- No Spurious Local Minima in Nonconvex Low Rank Problems: A Unified Geometric Analysis
- Stochastic Gradient Descent Optimizes Over-parameterized Deep ReLU Networks
- Qualitatively characterizing neural network optimization problems
- Essentially No Barriers in Neural Network Energy Landscape
- The loss surface of deep and wide neural networks
- Spurious Local Minima are Common in Two-Layer ReLU Neural Networks
- On the loss landscape of a class of deep neural networks with no bad local valleys
- Optimization Landscape and Expressivity of Deep CNNs
- Understanding the Loss Surface of Neural Networks for Binary Classification
- Spurious Valleys in Two-layer Neural Network Optimization Landscapes
- Adding One Neuron Can Eliminate All Bad Local Minima
- A mean-field limit for certain deep neural networks
- Mean Field Limit of the Learning Dynamics of Multilayer Neural Networks
- A Generalized Neural Tangent Kernel Analysis for Two-layer Neural Networks
- Topology and Geometry of Half-Rectified Network Optimization
- A Note on the Global Convergence of Multilayer Neural Networks in the Mean Field Regime