On Lipschitz Regularization of Convolutional Layers using Toeplitz Matrix Theory
This addresses the need for efficient and accurate Lipschitz regularization to improve training stability, generalization, and robustness in deep learning, representing a novel method for a known bottleneck.
The paper tackled the problem of Lipschitz regularization in Convolutional Neural Networks by introducing a new upper bound for convolutional layers that is both tight and easy to compute, leading to an algorithm for training such networks.
This paper tackles the problem of Lipschitz regularization of Convolutional Neural Networks. Lipschitz regularity is now established as a key property of modern deep learning with implications in training stability, generalization, robustness against adversarial examples, etc. However, computing the exact value of the Lipschitz constant of a neural network is known to be NP-hard. Recent attempts from the literature introduce upper bounds to approximate this constant that are either efficient but loose or accurate but computationally expensive. In this work, by leveraging the theory of Toeplitz matrices, we introduce a new upper bound for convolutional layers that is both tight and easy to compute. Based on this result we devise an algorithm to train Lipschitz regularized Convolutional Neural Networks.