Wasserstein distributional robustness of neural networks by Bai, X, He, G, Jiang, Y, Obłój, J

“…Deep neural networks are known to be vulnerable to adversarial attacks (AA).For an image recognition task, this means that a small perturbation of the original can result in the image being misclassified.Design of such attacks as well as methods of adversarial training against them are subject of intense research.We re-cast the problem using techniques of Wasserstein distributionally robust optimization (DRO) and obtain novel contributions leveraging recent insights from DRO sensitivity analysis.We consider a set of distributional threat models.Unlike the traditional pointwise attacks, which assume a uniform bound on perturbation of each input data point, distributional threat models allow attackers to perturb inputs in a nonuniform way.We link these more general attacks with questions of out-of-sample performance and Knightian uncertainty.To evaluate the distributional robustness of neural networks, we propose a first-order AA algorithm and its multistep version.Our attack algorithms include Fast Gradient Sign Method (FGSM) and Projected Gradient Descent (PGD) as special cases.Furthermore, we provide a new asymptotic estimate of the adversarial accuracy against distributional threat models.The bound is fast to compute and first-order accurate, offering new insights even for the pointwise AA.It also naturally yields out-of-sample performance guarantees.We conduct numerical experiments on CIFAR-10, CIFAR-100, ImageNet datasets using DNNs on RobustBench to illustrate our theoretical results.Our code is available at https://github.com/JanObloj/W-DRO-Adversarial-Methods.…”

Lagrangian decomposition for neural network verification by Bunel, R, De Palma, A, Desmaison, A, Dvijotham, K Dj, Kohli, P, Torr, PHS, Kumar, MP

“…A fundamental component of neural network verification is the computation of bounds on the values their outputs can take. …”

Towards efficient and reliable neural networks by de Jorge Aranda, P

“…<p>Deep neural networks have achieved remarkable results in various applications, but when deployed to autonomous agents in the real world several challenges arise. …”

Adversarial robustness of Bayesian neural networks by Wicker, M

“…<p>This thesis puts forward methods for computing local robustness of probabilistic neural networks, specifically those resulting from Bayesian inference. …”

Graph neural networks for network analysis by He, Y

Subjects: “…Neural networks (computer science)…”

Formal synthesis of Lyapunov neural networks by Abate, A, Ahmed, D, Giacobbe, M, Peruffo, A

“…Our method supports neural networks with polynomial activation functions and multiple depth and width, which display wide learning capabilities. …”

Analysis of robust neural networks for control by Newton, M

Subjects:

Learning invariant representations in neural networks by Fuchs, FB

“…The presented work focuses on invariant and equivariant neural network layers, putting symmetries at the centre of neural network architecture design. …”

Spoken letter recognition with neural networks by Reynolds, J, James Henry Reynolds

Subjects:

Probabilistic safety for bayesian neural networks by Wicker, M, Laurenti, L, Patane, A, Kwiatkowska, M

“…We study probabilistic safety for Bayesian Neural Networks (BNNs) under adversarial input perturbations. …”

Ontology reasoning with deep neural networks by Hohenecker, P, Lukasiewicz, T

“…In this paper, we employ state-of-the-art methods for training deep neural networks to devise a novel model that is able to learn how to effectively perform logical reasoning in the form of basic ontology reasoning. …”

TAILORING THE PERFORMANCE OF ATTRACTOR NEURAL NETWORKS by Wong, K, Sherrington, D

PATTERN SELECTIVITY IN OPTIMIZED NEURAL NETWORKS by Rau, A, Wong, K, Sherrington, D

THE OPTIMAL RETRIEVAL IN BOOLEAN NEURAL NETWORKS by Wong, K, Sherrington, D

Orbit-equivariant graph neural networks by Morris, M, Grau, BC, Horrocks, I

“…Equivariance is an important structural property that is captured by architectures such as graph neural networks (GNNs). However, equivariant graph functions cannot produce different outputs for similar nodes, which may be undesirable when the function is trying to optimize some global graph property. …”

Exploiting sparsity for neural network verification by Newton, M, Papachristodoulou, A

“…The problem of verifying the properties of a neural network has never been more important. This task is often done by bounding the activation functions in the network. …”

STORAGE PROPERTIES OF BOOLEAN NEURAL NETWORKS by Wong, K, Sherrington, D

Individual fairness guarantees for neural networks by Benussi, E, Patane, A, Wicker, M, Laurenti, L, Kwiatkowska, M

“…We consider the problem of certifying the individual fairness (IF) of feed-forward neural networks (NNs). In particular, we work with the epsilon-delta-IF formulation, which, given a NN and a similarity metric learnt from data, requires that the output difference between any pair of epsilon-similar individuals is bounded by a maximum decision tolerance delta >= 0. …”

Encryption function on artificial neural network by Al Azawee, H., Husien, S., Yunus, M.A.M.

“…In this paper, we try to decrypt automatically using artificial neural network by decryption through multilayer perceptron and radial basis function; networks were tested using the interface by calculating the error rates of decrypted images.…”

Backpropagation Neural Network For Colour Recognition by AL-Naqeeb, Abdul Aziz Hussien

“…It is found that RGB is useful when used with Neural Network and the Normalized RGB value is faster in the learning of neural network.…”
Get full text