Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

LDLT $\mathcal{L}$-Lipschitz Network: Generalized Deep End-To-End Lipschitz Network Construction

About

Deep residual networks (ResNets) have demonstrated outstanding success in computer vision tasks, attributed to their ability to maintain gradient flow through deep architectures. Simultaneously, controlling the Lipschitz constant in neural networks has emerged as an essential area of research to enhance adversarial robustness and network certifiability. This paper presents a rigorous approach to the general design of $\mathcal{L}$-Lipschitz deep residual networks using a Linear Matrix Inequality (LMI) framework. Initially, the ResNet architecture was reformulated as a cyclic tridiagonal LMI, and closed-form constraints on network parameters were derived to ensure $\mathcal{L}$-Lipschitz continuity; however, using a new $LDL^\top$ decomposition approach for certifying LMI feasibility, we extend the construction of $\mathcal{L}$-Lipchitz networks to any other nonlinear architecture. Our contributions include a provable parameterization methodology for constructing Lipschitz-constrained residual networks and other hierarchical architectures. Cholesky decomposition is also used for efficient parameterization. These findings enable robust network designs applicable to adversarial robustness, certified training, and control systems. The $LDL^\top$ formulation is shown to be a tight relaxation of the SDP-based network, maintaining full expressiveness and achieving 3\%-13\% accuracy gains over SLL Layers on 121 UCI data sets.

Marius F.R. Juston, Ramavarapu S. Sreenivas, Dustin Nottage, Ahmet Soylemezoglu• 2025

Related benchmarks

TaskDatasetResultRank
Mean Accuracy121 benchmarks tasks
Wins A94
15
Certified AccuracyMean Certified Accuracy (72/255) (test)
Wins A83
15
Certified RobustnessMean Certified Accuracy 36/255
Wins A81
15
Certified RobustnessMean Certified Accuracy 108/255
Wins A83
15
Certified AccuracyMean Certified Accuracy 255/255
Total Samples (N)121
15
Classification121 UCI data sets mean across
Accuracy72.23
6
Classification121 UCI data sets
Average Rank2.434
6
Certified AccuracyUCI radius 255/255 overall comparison
Average Rank2.55
6
Mean Certified AccuracyUCI data sets (test)
Average Rank2.715
6
Mean Certified AccuracyUCI 108/255 radius (test)
Average Rank2.562
6
Showing 10 of 11 rows

Other info

Follow for update