Method detects adversarial samples using influence functions and nearest neighbors.
problem Detecting adversarial attacks on deep neural networks.
method Influence functions and k-NN model on activation layers.
result Successfully distinguishes adversarial examples with state-of-the-art results.
Paper introduces AIF to analyze robust optimization effects.
problem Quantifying robust optimization's impact on model optimizers and losses.
method Inspired by robust statistics, AIF is introduced to measure model sensitivity.
result AIF reveals how model complexity and randomized smoothing affect model sensitivity.
The paper examines how to make multivariate estimators robust against adversarial data modifications.
problem Adversarial robustness of multivariate M-Estimators. method Adversarial Influence Function (AIF) to measure robustness; optimal modification strategy and AIF characterization; optimal M-estimator design. result Characterization of optimal M-estimator with smallest AIF for joint location and scale estimation. We consider the problem of influence maximization in fixed networks for contagion models in an adversarial setting. The goal is to select an optimal set of nodes to seed the influence process, such that the number of influenced nodes at the conclusion of the campaign is as large as possible. We formulate the problem as…
mFI-PSO generates effective adversarial images for DNNs.
problem Vulnerability of DNNs to small perturbations in images.
method Uses mFI for pixel selection and PSO for objective functions.
result mFI-PSO effectively designs flexible adversarial images.
I-GCN improves GCNs' robustness against adversarial attacks.
problem Adversarial attacks degrade GCNs' performance in security-critical applications.
method Influence mechanism divides node effects into introverted and extroverted influences.
result I-GCN achieves higher accuracy rates than state-of-the-art methods in defending against adversarial attacks.
Enhances influence functions for deep models without costly Hessian inversion.
problem Challenges in computing Hessian matrix inverse for deep models.
method Substitutes Hessian inverse with identity matrix and extends applications.
result TracIn method performs well in measuring model utility and fairness.
Adversarial attacks on probabilistic state-space models affect latent state and policy decisions.
problem Robust reinforcement learning under adversarial observability.
method Analyzing adversarial attacks on linear probabilistic state-space models.
result Demonstrating the influence of adversarial observations on latent state and policy decisions.
Analyzes adversarial training's impact on loss landscape, proposing PAS to improve model performance.
problem Challenges in optimizing models under adversarial training due to loss landscape properties.
method Analytical studies of adversarial loss functions, numerical analyses, PAS strategy.
result Adversarial training impairs optimization, but PAS strategy improves model performance.
New method detects anomalies in systems influenced by their environment.
problem Detecting anomalies in systems under environmental influence.
method Adversarial learning and time series representation learning.
result Successfully addresses label sparsity and subjectivity in anomaly detection.
New method for fair influence maximization in social networks.
problem Maximizing influence while ensuring fairness across sensitive attributes.
method Co-training an auto-encoder and discriminator to create fair graph embeddings.
result Our method reduces disparity while maintaining competitive influence maximization performance.
Data augmentation impacts adversarial risk; careful application recommended.
problem Understanding how data augmentation affects adversarial risk in deep learning.
method Empirical analysis using three measures of adversarial risk.
result Data augmentation does not always improve adversarial risk; augmented data influences models more.
State-of-the-art deep neural networks are known to be vulnerable to adversarial examples, formed by applying small but malicious perturbations to the original inputs. Moreover, the perturbations can \textit{transfer across models}: adversarial examples generated for a specific model will often mislead other unseen mode…
Proposes a physics-informed VAE for disentangling physics from confounding influences.
problem Challenges in inferring and predicting physical systems under partial knowledge.
method Physics-informed variational autoencoder with adversarial training.
result Model successfully disentangles known physics from confounding influences.
Adversarially-trained models transfer better in limited data scenarios.
problem Improving transfer learning performance with limited data.
method Adversarially-train deep nets, freeze early layers, fine-tune last layers, observe shape vs texture bias.
result Adversarially-trained models transfer better than non-adversarially-trained models, especially with limited data.
Study reveals how features influence deep network decision boundaries.
problem Understanding the role of features in neural network decision boundaries.
method Adopted adversarial robustness tools to measure changes in CNN decision boundaries.
result Neural networks exhibit high invariance to non-discriminative features and are sensitive to small perturbations of training samples.
This study evaluates adversarial attacks and defenses for chest X-ray disease classification.
problem Vulnerability of deep neural networks to adversarial examples in chest X-ray disease detection.
method Detailed introduction and evaluation of various attack and defense methods.
result Attack and defense methods perform poorly with excessive iterations and large perturbations.
Study shows how to manipulate VAEs for attacks and assess their robustness.
problem Adversarial attacks on Variational Autoencoders (VAE).
method Examine modifications to VAEs and propose metrics for robustness.
result Metrics to quantify the robustness of VAEs to adversarial attacks.
New algorithm for linear optimization with adaptive corruption.
problem Stochastic linear optimization under adversarial corruption.
method Algorithm uses Löwner-John's ellipsoid for exploration and divides time into epochs.
result Regret increases linearly with corruption amount.
Framework for balancing accuracy and robustness in machine learning.
problem Balancing accuracy and robustness in machine learning models.
method Developed a framework and introduced quantities to characterize the trade-off, including a simple trade-off curve and an influence function.
result Theoretical insight and experimental demonstration of the trade-off between accuracy and robustness.
ANTIDOTE reduces noisy labels influence during learning.
problem Learning with noisy labels.
method Information-divergence neighborhood relaxation and adversarial training.
result ANTIDOTE outperforms standard cross-entropy loss in noisy label settings.
Machine learning models are vulnerable to adversarial examples: small changes to images can cause computer vision models to make mistakes such as identifying a school bus as an ostrich. However, it is still an open question whether humans are prone to similar mistakes. Here, we address this question by leveraging recen…
Defense against user shilling attacks in collaborative filtering using edge reweighting.
problem Vulnerability of collaborative filtering to profile injection attacks.
method Adversarial robustness based edge reweighting to attenuate non-robust edges.
result Effective defense against various types of attacks demonstrated through experiments.
Influence functions are inaccurate in deep learning models, especially for deeper networks.
problem Inaccuracies in influence functions in deep learning models.
method Empirical study of influence functions in neural network models trained on various datasets.
result Influence estimates are often erroneous for deeper networks and require regularization.
New algorithm eliminates sign function in PGD attacks, improving performance.
problem Improving robustness of neural networks against adversarial attacks.
method Proposes a new raw gradient descent (RGD) algorithm that eliminates the sign function in PGD attacks.
result The RGD algorithm outperforms PGD and other competitors in various settings.
New research shows adversarial defenses increase risk of membership inference attacks.
problem Combining security and privacy domains to measure adversarial defenses' impact on membership inference attacks.
method Measuring the success of membership inference attacks against six state-of-the-art adversarial defense methods.
result Adversarial defense methods can increase the risk of membership inference attacks.
Adaptive methods produce less robust models to adversarial examples.
problem Robustness of machine learning models to adversarial examples.
method Comparison of adaptive and non-adaptive optimization methods in linear regression.
result Non-adaptive methods consistently produce more robust models.
RelatIF selects more intuitive training examples for explaining model predictions.
problem Influence functions identify outliers as explanatory examples, leading to poor explanations.
method RelatIF separates global and local influence, optimizing for local relative to global effects.
result Examples selected by RelatIF are more intuitive than those from influence functions.
Establishes statistical and computational bounds for influence diagnostics.
problem Identifying influential datapoints or subsets in machine learning models.
method Finite-sample statistical bounds and computational complexity for influence functions and approximate maximum influence perturbations.
result Established statistical and computational guarantees for influence diagnostics.
Influence functions help study large language model generalization, revealing surprising decay patterns.
problem Understanding and mitigating risks in large language models (LLMs).
method Eigenvalue-corrected Kronecker-Factored Approximation (EK-FAC) to scale influence functions to LLMs.
result Influences decay to near-zero when key phrases order is flipped, revealing a surprising limitation.
Recently, there has been an abundance of works on designing Deep Neural Networks (DNNs) that are robust to adversarial examples. In particular, a central question is which features of DNNs influence adversarial robustness and, therefore, can be to used to design robust DNNs. In this work, this problem is studied throug…
New research investigates why influence functions are fragile and proposes new validation procedures.
problem Understanding and mitigating the fragility of influence functions in deep learning model explanations.
method Verification of influence functions using various conditions and procedures, including convexity and non-convexity.
result Validation procedures may cause the observed fragility of influence functions.
Proposes second-order influence functions for identifying influential groups in test-time predictions.
problem Identifying influential groups in test-time predictions for black-box models.
method Second-order approximations of the effect of removing a group of training samples on model predictions.
result Improves the correlation between computed influence values and ground truth values for linear models.
Study shows influence functions are poor for neural networks but useful for identifying influential examples.
problem Influence functions misalign with leave-one-out retraining in neural networks.
method Decomposed the discrepancy into five terms and studied their contributions across different architectures and datasets.
result Influence functions are a good approximation to the proximal Bregman response function (PBRF), useful for identifying influential examples.
Study on adversarial examples from data size, task, and model factors.
problem Understanding adversarial examples from data size, task, and model perspectives.
method Systematic study on adversarial examples from three aspects: data size, task-dependent, and model-specific factors.
result Adversarial generalization requires more data than standard generalization.
Paper introduces a multi-stage influence function to track model predictions.
problem Improving natural language processing and computer vision performance.
method Develops a multi-stage influence function to track predictions from finetuned models back to pretraining data.
result Identifies pretraining examples contributing most to finetuning task predictions.
The paper extends influence functions to sequence tagging tasks for better model interpretability.
problem Lack of interpretability methods for sequence tagging models.
method Define and compute influence of training instance segments on test segment predictions.
result The segment influence method tracks with true influence and identifies annotation errors.
Linearized attention fails to converge to NTK limit even at large widths.
problem Understanding the convergence of attention mechanisms to the kernel regime.
method Analyzes linearized attention and its relationship to the NTK limit, considering practical widths and conditions.
result Linearized attention does not converge to its NTK limit at any practical width, revealing a fundamental trade-off.
Better Hessian approximations improve influence function attributions in deep learning.
problem Influence functions are difficult to compute due to ill-conditioned Hessians, leading to poor data attribution performance.
method Investigated the impact of Hessian approximation quality on influence-function attributions in a controlled setting.
result Better Hessian approximations consistently yield better influence score quality.
Ensemble of diverse CNNs detects and mitigates adversarial attacks.
problem Detecting and defending against adversarial attacks.
method An ensemble of specialized CNNs with a voting mechanism.
result Significant reduction in adversarial attack risk rate.
The paper evaluates machine learning cyber defenses using log data against adversarial attacks.
problem Evaluating the robustness of machine learning cyber defenses against adversarial attacks.
method Developed a testing framework using deep reinforcement learning and adversarial natural language processing.
result Higher dropout levels increase robustness, with 90% dropout probability showing the highest robustness.
Despite the great success achieved in machine learning (ML), adversarial examples have caused concerns with regards to its trustworthiness: A small perturbation of an input results in an arbitrary failure of an otherwise seemingly well-trained ML model. While studies are being conducted to discover the intrinsic proper…
Framework detects and mitigates data-poisoning attacks in causal effect estimation.
problem Vulnerability to append-only attacks in observational causal analyses.
method Develops a data-poisoning audit for augmented inverse-probability-weighted estimation.
result Proposes a greedy scan to compute exact worst-case movement at every append budget.
There is an especially strong need in modern large-scale data analysis to prioritize samples for manual inspection. For example, the inspection could target important mislabeled samples or key vulnerabilities exploitable by an adversarial attack. In order to solve the "needle in the haystack" problem of which samples t…
The paper simplifies influence computations for large-scale machine learning models.
problem Improving training efficiency and accuracy in large-scale models.
method Study influence functions, define memorization, simplify computations.
result Influence functions can be practical for large-scale models, indicating memorization.
StealthRank subtly boosts LLM rankings without detectable anomalies.
problem Adversarial manipulation of LLM-driven ranking systems.
method Energy-based optimization with Langevin dynamics for stealthy prompt generation.
result StealthRank outperforms existing methods in covertly boosting rankings.
Deep neural networks are vulnerable to adversarial examples. Prior defenses attempted to make deep networks more robust by either changing the network architecture or augmenting the training set with adversarial examples, but both have inherent limitations. Motivated by recent research that shows outliers in the traini…
New IF method improves accuracy in deep neural networks with noisy data.
problem Inaccurate influence estimates in deep neural networks, especially with noisy data.
method Established a connection between influence estimation error, validation set risk, and sharpness, introducing a novel estimation form for flat validation minima.
result Our novel Influence Function approach provides more accurate influence estimates, validated across various tasks.