Bayesian adversaries can outsmart traditional adversarial attacks.
problem Bayesian adversaries can manipulate machine learning models through small perturbations.
method Developed a continuous-time particle system (Abram) to approximate the gradient flow of the Bayesian adversarial robustness problem.
result Abram approximates the minimizer of the Bayesian adversarial robustness problem under certain assumptions.
New reformulations for multiclass classification problems using optimal transport.
problem Adversarial multiclass classification problems.
method Multimarginal optimal transport formulation.
result Reveals geometric structure and extends binary classification results.
Adversarial examples are maliciously tweaked images that can easily fool machine learning techniques, such as neural networks, but they are normally not visually distinguishable for human beings. One of the main approaches to solve this problem is to retrain the networks using those adversarial examples, namely adversa…
Classification problems in security settings are usually contemplated as confrontations in which one or more adversaries try to fool a classifier to obtain a benefit. Most approaches to such adversarial classification problems have focused on game theoretical ideas with strong underlying common knowledge assumptions, w…
Adversarial training improves linear regression solutions, revealing sparsity and abrupt interpolation.
problem Adversarial attacks on linear regression models.
method Formulated as a convex problem, adversarial training is used to find robust solutions that are sparse and interpolate data.
result Adversarial training with small disturbances gives the solution with the minimum-norm that interpolates the training data, revealing abrupt transition into interpolation.
Here we propose a general theoretical method for analyzing the risk bound in the presence of adversaries. Specifically, we try to fit the adversarial learning problem into the minimax framework. We first show that the original adversarial learning problem can be reduced to a minimax statistical learning problem by intr…
Adversarial training provides a principled approach for training robust neural networks. From an optimization perspective, adversarial training is essentially solving a bilevel optimization problem. The leader problem is trying to learn a robust classifier, while the follower problem is trying to generate adversarial s…
Study proves existence of robust classifiers in multiclass adversarial training.
problem Proves existence of robust classifiers in multiclass adversarial training.
method Three models of adversarial training in multiclass classification, proving existence of Borel measurable robust classifiers.
result Proves existence of Borel measurable robust classifiers in each model.
A wide range of defenses have been proposed to harden neural networks against adversarial attacks. However, a pattern has emerged in which the majority of adversarial defenses are quickly broken by new attacks. Given the lack of success at generating robust defenses, we are led to ask a fundamental question: Are advers…
New algorithm optimally identifies best arm in both stochastic and adversarial settings.
problem Best arm identification in stochastic and adversarial reward scenarios.
method Parameter-free algorithm designed to be robust to adversarial rewards and optimal in stochastic problems.
result Algorithm's error rate matches optimal bounds in stochastic problems and is robust to adversarial rewards.
ALPS improves neural network robustness and generalization.
problem Challenges in designing effective regularization schemes for adversarial robustness.
method Adversarial Labelling of Perturbed Samples (ALPS) using synthetic samples and min-max formulation.
result ALPS achieves state-of-the-art regularization performance and adversarial robustness.
Linear classifiers can resist adversarial attacks on Gaussian data.
problem Adversarial attacks on high-dimensional data.
method Adversarial training of linear classifiers on Gaussian data.
result Linear classifiers can resist adversarial attacks on Gaussian data.
Optimizes deep learning training by treating it as an optimal control problem.
problem Fragility of deep neural networks to adversarial inputs.
method Formulates adversarial training as a min-max optimization problem and interprets it as an optimal control problem.
result Provides the first convergence analysis of adversarial training algorithm.
Study stabilizes adversarial training in neural networks over infinite-dimensional spaces.
problem Stability issues in adversarial training of neural networks.
method Functional analysis of minimax optimization over infinite-dimensional spaces of continuous functions and probability measures.
result Convergence property of minimax problems under certain conditions, interpreted as stabilization techniques.
A new method reduces adversarial training time without overfitting.
problem Catastrophic overfitting in single-step adversarial training.
method FGSMPR: FGSM with PGD Regularization.
result Reduces the gap to multi-step adversarial training.
New geometric insights reveal properties of adversarial training problems.
problem Adversarial training in binary classification.
method Equivalence with regularized risk minimization and convex relaxations.
result Existence of minimal and maximal solutions, and regular solutions.
Adversarial examples are inputs to machine learning models designed to cause the model to make a mistake. They are useful for understanding the shortcomings of machine learning models, interpreting their results, and for regularisation. In NLP, however, most example generation strategies produce input text by using kno…
Efficient algorithms speed up adversarial training for linear models.
problem Adversarial training for linear models is computationally expensive.
method Tailored optimization algorithms for regression and classification.
result Significantly faster convergence rates for large-scale problems.
Numerous methods for crafting adversarial examples were proposed recently with high success rate. Since most existing machine learning based classifiers normalize images into some continuous, real vector, domain firstly, attacks often craft adversarial examples in such domain. However, "adversarial" examples may become…
Paper tackles online allocation problems using adversarial training.
problem Online bipartite matching, especially in AdWords.
method Constructs a framework combining game theory, adversarial training, and GANs.
result Designs robust algorithms that perform well under practical and adversarial conditions.
Study adversarial classification with data corruption up to ε, deriving geometric flows.
problem Optimizing classifiers against adversarial data corruption.
method Variational analysis, geometric flows, mean curvature equations.
result Rigorous proof of initial value problem for small ε, global minimizer of adversarial problem.
SPAT improves adversarial robustness by preserving semantics in adversarial training.
problem Adversarial examples often have different semantics than original data, introducing unintended biases.
method Semantics-preserving adversarial training (SPAT) that encourages pixel perturbation shared among all classes.
result SPAT improves adversarial robustness and achieves state-of-the-art results in CIFAR-10 and CIFAR-100.
Study efficient learning of robust halfspaces with noise.
problem Learning robust halfspaces in the presence of adversarial perturbations and random label noise.
method Provides conditions for robust learnability and a simple algorithm for any ℓ_p perturbation.
result Simple computationally efficient algorithm for robust learning with random label noise.
The paper sets lower bounds for adversarial robustness in multiclass classification.
problem Adversarial robustness in multiclass classification with arbitrary loss functions.
method Dual and barycentric reformulations for robust risk minimization.
result Sharp lower bounds for adversarial risks are computed efficiently.
New method simplifies and improves imitation learning without adversarial techniques.
problem Stable optimization and convergence issues in adversarial imitation learning methods.
method Proposes a non-adversarial framework for imitation learning, providing stronger convergence guarantees.
result Shows AIRL as a special case and derives new algorithms for offline imitation learning.
We study the stochastic multi-armed bandits problem in the presence of adversarial corruption. We present a new algorithm for this problem whose regret is nearly optimal, substantially improving upon previous work. Our algorithm is agnostic to the level of adversarial contamination and can tolerate a significant amount…
Machine learning models are vulnerable to adversarial inputs that induce seemingly unjustifiable errors. As automated classifiers are increasingly used in industrial control systems and machinery, these adversarial errors could grow to be a serious problem. Despite numerous studies over the past few years, the field of…
Although Deep Neural Networks(DNNs) have achieved successful applications in many fields, they are vulnerable to adversarial examples.Adversarial training is one of the most effective methods to improve the robustness of DNNs, and it is generally considered as solving a saddle point problem that minimizes risk and maxi…
New algorithms improve online prediction from experts with privacy constraints.
problem Online prediction from experts under privacy constraints.
method Proposed and analyzed new algorithms for approximate and pure differential privacy.
result Achieved improved regret bounds for various adversaries.
A new method for estimating adversarial strategies in nonlinear systems.
problem Inferring an intelligent adversarial agent's strategy in highly nonlinear systems.
method Formulated inverse cognition as a nonlinear Gaussian state-space model and developed an inverse UKF (IUKF) system.
result The estimation error of IUKF converges and closely follows the recursive Cramér-Rao lower bound.
We provide a methodology, resilient feature engineering, for creating adversarially resilient classifiers. According to existing work, adversarial attacks identify weakly correlated or non-predictive features learned by the classifier during training and design the adversarial noise to utilize these features. Therefore…
The paper improves smoothed analysis for online problems with adaptive adversaries.
problem Online prediction, discrepancy minimization, and online optimization with adaptive adversaries.
method General technique to prove smoothed guarantees against adaptive adversaries, reducing to simpler oblivious adversaries.
result Strong smoothed guarantees for three online problems, matching or improving previous results.
Many machine learning models are vulnerable to adversarial attacks; for example, adding adversarial perturbations that are imperceptible to humans can often make machine learning models produce wrong predictions with high confidence. Moreover, although we may obtain robust models on the training dataset via adversarial…
Adversarial robustness improved by abstaining from decisions.
problem Improving classification accuracy in the presence of adversarial perturbations.
method Introducing an abstain option in binary classification problems, using metrics to quantify performance and robustness.
result There is a tradeoff between nominal performance and adversarial robustness.
We investigate the non-identifiability issues associated with bidirectional adversarial training for joint distribution matching. Within a framework of conditional entropy, we propose both adversarial and non-adversarial approaches to learn desirable matched joint distributions for unsupervised and supervised tasks. We…
AdvImmune improves certifiable robustness of GNNs against adversarial attacks.
problem Vulnerability of graph neural networks to adversarial attacks.
method Proposes AdvImmune, an algorithm that optimizes with meta-gradient to improve certifiable robustness.
result Remarkably improves the ratio of robust nodes by 12%, 42%, 65% with an affordable immune budget of only 5% edges.
Deep learning models, which are increasingly being used in the field of medical image analysis, come with a major security risk, namely, their vulnerability to adversarial examples. Adversarial examples are carefully crafted samples that force machine learning models to make mistakes during testing time. These maliciou…
Paper improves adversarial training using a learned optimizer.
problem Improving robustness of deep learning models against adversarial attacks.
method Empirically identified PGD attack's limitations and used a learning-to-learn framework to train an adaptive inner optimizer.
result The proposed framework consistently improves model robustness over traditional adversarial training methods.
Survey of GANs and autoencoders, addressing mode collapse and likelihood issues.
problem Addressing mode collapse and likelihood issues in GANs and autoencoders.
method Explains various GAN and autoencoder variants, their applications, and methods to resolve issues.
result Various methods to resolve mode collapse and improve likelihood in GANs and autoencoders.
Analyzes biased random walks and corrupted intervals in adversarial settings.
problem Learning thresholds and intervals in adversarial conditions.
method Analyzes biased random walks and corrupted intervals under adversarial design.
result Analyzes the expected behavior of biased random walks and corrupted intervals.
Adversarial training improves linear regression solutions, offering robustness against small perturbations.
problem Vulnerability of linear models to adversarial perturbations.
method Formulated as a min-max problem, adversarial training minimizes the best solution under worst-case attacks.
result Adversarial training yields the minimum-norm interpolating solution in overparameterized models, equivalent to parameter shrinking methods in underparameterized models.
New method UADs improves transferability of adversarial perturbations.
problem Transferability of adversarial perturbations across different DNN architectures.
method Proposes Universal Adversarial Directions (UADs) to improve transferability.
result UADs can achieve a Nash equilibrium, indicating potential transferability.
We derive bounds for a notion of adversarial risk, designed to characterize the robustness of linear and neural network classifiers to adversarial perturbations. Specifically, we introduce a new class of function transformations with the property that the risk of the transformed functions upper-bounds the adversarial r…
Paper tackles robustness in adversarial noise with a meta-optimizer.
problem Sensitivity to adversarial noise hinders machine learning deployment.
method Meta-optimizer learns to robustly optimize models using adversarial examples.
result Meta-optimizer transfers adversarial knowledge to new models without generating new examples.
Optimal adversarial attacks minimize mutual information, revealing classifier vulnerabilities.
problem Designing optimal attacks to degrade machine learning performance.
method Information-theoretic approach to finding optimal perturbations.
result Optimal attacks minimize mutual information between degraded and original signals.
New bounds show complexity of adversarial decision making.
problem Understanding sample efficiency in adversarial decision making.
method New upper and lower bounds on Decision-Estimation Coefficient.
result Decision-Estimation Coefficient is necessary and sufficient for low regret in adversarial decision making.
A Markov Decision Process (MDP) is a popular model for reinforcement learning. However, its commonly used assumption of stationary dynamics and rewards is too stringent and fails to hold in adversarial, nonstationary, or multi-agent problems. We study an episodic setting where the parameters of an MDP can differ across…
New adversarial training method improves robustness of power system controllers.
problem Designing robust controllers for complex cyber-physical power systems.
method Adversarial training approach with fixed opponent policy.
result Adversarial trained controllers show useful preventive behaviors in the N-1 problem.