Convergence of gradient flow for learning convolutional neural networks
By: Jona-Maria Diederen, Holger Rauhut, Ulrich Terstiege
Convolutional neural networks are widely used in imaging and image recognition. Learning such networks from training data leads to the minimization of a non-convex function. This makes the analysis of standard optimization methods such as variants of (stochastic) gradient descent challenging. In this article we study the simplified setting of linear convolutional networks. We show that the gradient flow (to be interpreted as an abstraction of gradient descent) applied to the empirical risk defined via certain loss functions including the square loss always converges to a critical point, under a mild condition on the training data.
Similar Papers
Gradient Flow Equations for Deep Linear Neural Networks: A Survey from a Network Perspective
Machine Learning (CS)
Helps computers learn by simplifying math.
Convergence of gradient based training for linear Graph Neural Networks
Machine Learning (CS)
Makes computer learning on networks faster.
On the (almost) Global Exponential Convergence of the Overparameterized Policy Optimization for the LQR Problem
Optimization and Control
Makes computer learning faster and better.