Sep 2026· Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence· 0 citations· 61 references
TL;DR
It is demonstrated that over-parameterization can be exploited not merely to enhance generalization, but also to mitigate neuron instability, one of the parameters affecting the efficiency of verification.
Abstract
Over-parameterized neural networks, i.e., models with excess
capacity that can fit training data exactly, have demonstrated superior
generalization performance compared to classical models with balanced
capacity. Nevertheless, their deployment in safety-critical domains re-
mains severely constrained by their susceptibility to, e.g., natural per-
turbations and adversarial manipulations. Verification techniques can
solve such problems, but the computational cost of these methods of-
ten scales poorly, specifically when applied to large models. In this work,
we demonstrate that over-parameterization can be exploited not merely
to enhance generalization, but also to mitigate neuron instability, one of
the parameters affecting the efficiency of verification. Our experimental
findings suggest that over-parameterization may serve as a crucial mech-
anism for reconciling the long-standing trade-off between generalization
and verifiability of neural networks.
This study shows that template subsumption rates can vary widely across scenarios, and presents FastCert, a novel technique for automatically distributing templates across neural network layers to increase performance impact, eschewing templates entirely if they are unlikely to produce a speedup.
Kanak Das, Shubham Ugare, B. E. Chang et al.· 0 citations
Mutual Heterogeneous Learning (MHL) is proposed, a framework enabling robust pruning via single-model inference that significantly outperforms single-model baselines in both adversarial robustness and corruption robustness, while maintaining competitive clean accuracy.
Jin-Hui Yu, Zikai Zhang, Khaled A. Harras et al.· Proceedings of the Thirty-Fi...· 0 citations
A novel SNN named arbitrary distribution injection is proposed, which enables distribution modeling from nonpredefined, data-dependent distributions and introduces a conditional stochastic feature mapping mechanism to model feature distributions, together with a theoretically grounded variance-regularization loss.
Rui Zhou, Hao Yang, Wen-Xu Wang et al.· IEEE Transactions on Neural...· 0 citations
Across CIFAR10, COIL100, and ETH80 under FGSM and Genetic Attack, ViT + Classical HDC degrades more gracefully than Pure HDC and ViT + HDC baselines, and achieves higher normalized AURC, lower attack-time decision margins, and larger gains from partial adversarial retraining.
Hamza Errahmouni Barkam, Salaar Saraj, Zhen Ye et al.· International Symposium on L...· 0 citations
It is shown how an adversarial model trainer can plant backdoors in a large class of deep, feedforward neural networks, demonstrating a fundamental power asymmetry between model trainers and model users.
Andrej Bogdanov, Alon Rosen, N. Vafa· IACR Cryptology ePrint Archi...· 1 citation
This paper investigates the certified and empirical robustness of various Kolmogorov-Arnold network architectures against strong evasion attacks and systematically evaluates the robustness of various defended and undefended KAN models under FGSM, PGD, and C&W attacks.
Mohammad Meymani, R. Razavi-Far· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
