Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

169,051 papers · 148 categories

Trend · papers per month

15.1%30.1%45.2%60.3% · Jun 202019922001200920182026
48 results for Machine Learning Defenses

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.

Interpreting machine learning models helps understand adversarial attacks and defenses.

problem Understanding model vulnerability to adversarial attacks.
method Model interpretation techniques to explore adversarial attacks and defenses.
result Interpretation methods can be applied to adversarial attacks and defenses.

A Bayesian framework models adversarial uncertainty for robust machine learning.

problem Vulnerability of machine learning models to adversarial attacks.
method Formal Bayesian framework that models adversarial uncertainty through a stochastic channel, articulating probabilistic assumptions.
result Explicitly modeling adversarial uncertainty leads to improved robustification strategies.

AttriGuard defends against attribute inference attacks via adversarial machine learning.

problem Attribute inference attacks on user data from public data.
method Two phases: finding minimum noise and sampling attribute values.
result AttriGuard outperforms existing methods in protecting user data.

Unified framework for analyzing model stealing attacks and defenses.

problem Vulnerability of ML applications to model stealing attacks.
method Developed a rigorous threat model and evaluation criteria, proposed methods to quantify attack and defense strategies.
result Demonstrated the importance of attack-specific perturbations for effective defenses.

New defense DMP preserves ML model utility while enhancing membership privacy.

problem Membership inference attacks (MIAs) on machine learning models.
method Knowledge distillation to train ML models with membership privacy.
result DMP provides significantly better tradeoffs between membership privacy and classification accuracies.

This study improves scalability of randomized smoothing for certifying classifier robustness.

problem Certifying machine learning classifiers against adversarial attacks is challenging and scalable solutions are needed.
method The study reviews and explores randomized smoothing and its derivatives, focusing on scalability.
result The study provides theoretical guarantees and discusses scalability challenges of randomized smoothing.

Surveying strategies for making machine learning models robust against adversarial attacks.

problem Ensuring machine learning models are robust and reliable in real-world applications.
method Taxonomy of adversarial attacks and defenses, Robust Optimization problem formulation, and survey of methods.
result Surveyed recent results in adversarial example generation, defense mechanisms, and formal robustness certificates.

Paper examines how adversarial ML attacks and defenses impact civil liberties and human rights.

problem Impact of adversarial machine learning on civil liberties and human rights.
method Uses insights from STS, anthropology, and human rights literature.
result Adversarial defenses can be used to suppress dissent and limit investigation of ML systems.

SPECTRE defends against backdoor attacks by amplifying corrupted data's spectral signature.

problem Backdoor attacks that change model behavior with specific triggers.
method Robust covariance estimation to amplify spectral signature of poisoned data.
result Clean model is completely removed from backdoor, even in hard-to-detect cases.

Machine learning is vulnerable to adversarial examples: inputs carefully modified to force misclassification. Designing defenses against such inputs remains largely an open problem. In this work, we revisit defensive distillation---which is one of the mechanisms proposed to mitigate adversarial examples---to address it…

2017-05-15abs ↗pdf ↗

Topology-aware generative models improve manifold-based defenses against adversarial examples.

problem Adversarial examples compromise the reliability of ML models, especially DNNs.
method Investigate if generative models used in manifold-based defenses need to be topology-aware.
result Topology-aware generative models enhance the robustness of manifold-based defenses.

Deep Partition Aggregation defends against poisoning attacks with provable certificates.

problem Adversarial poisoning attacks corrupt classifier test-time behavior.
method Deep Partition Aggregation (DPA) is an ensemble method using hash partitions and base models trained on these partitions.
result DPA can certify >= 50% of test images against over 500 poison image insertions on MNIST, and nine insertions on CIFAR-10.

Paper explains AI's vulnerability to small changes and proposes a defense.

problem Adversarial attacks on machine learning systems.
method Information-theoretic approach, drawing on communication theory.
result The proposed defense method detects classifier errors caused by small perturbations.

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.

Two quantization-based defenses improve CNN robustness against adversarial attacks.

problem Adversarial examples threaten machine learning, especially CNNs.
method Constant Quantization (CQ) and Trainable Quantization (TQ) to increase CNN robustness.
result Significant increase in classification accuracy of perturbed images.

Adversarial examples that fool machine learning models, particularly deep neural networks, have been a topic of intense research interest, with attacks and defenses being developed in a tight back-and-forth. Most past defenses are best effort and have been shown to be vulnerable to sophisticated attacks. Recently a set…

2018-02-09abs ↗pdf ↗

Defense against model stealing attacks by adding noise to class probabilities.

problem Protecting machine learning models from model stealing attacks.
method Adding noise to class probabilities to limit the success of attacks.
result Our defense can degrade the accuracy of the stolen model by at least 20% or require up to 64 times more queries.

TensorShield defends images from adversarial attacks using tensor decomposition.

problem Adversarial attacks on images can fool deep neural networks.
method Tensor decomposition to find low-rank approximations of images, reducing high-frequency perturbations.
result TensorShield outperforms existing methods like SLQ by 14% against FGSM attacks.

New attacks bypass data sanitization defenses, increasing model errors.

problem Data poisoning attacks corrupt machine learning models trained on external data.
method Developed three attacks that coordinate poisoned points and formulate as optimization problems.
result 3% poisoned data increases test error from 3% to 24% on Enron spam detection.

Research evaluates data poisoning attacks on regression learning and introduces a new defense strategy.

problem Data poisoning attacks on regression learning threaten model integrity in critical systems.
method Realistic scenarios, novel black-box attack, and evaluation on 26 datasets.
result Mean squared error (MSE) increases to 150% with only 2% poisoned samples.

This paper reviews defenses against adversarial learning attacks on statistical classifiers.

problem Adversarial attacks on machine learning systems, particularly statistical classifiers.
method Survey of recent work on test-time evasion, data poisoning, and reverse engineering attacks and defenses.
result Novel insights that challenge conventional AL wisdom and target unresolved issues.

The paper improves neural network robustness against multiple norm types of adversarial attacks.

problem Defending neural networks against adversarial attacks with different norms.
method Combining existing defense mechanisms to train neural networks robust against both \ell_\infty and 2\ell_2 attacks.
result New defense mechanisms offer better protection against both \ell_\infty and 2\ell_2 attacks.

Defense-GAN defends deep neural networks from adversarial attacks using generative models.

problem Vulnerability of deep neural networks to adversarial perturbations.
method Defense-GAN models the distribution of unperturbed images and finds a close output to a given image without adversarial changes.
result Defense-GAN consistently improves over existing defense strategies and works against various attack methods.

Explores security challenges of machine learning in real-world systems.

problem Vulnerabilities in machine learning models deployed in safety-critical systems.
method Broadens systems security view of ML vulnerabilities, identifies novel challenges, proposes mitigation suggestions.
result Highlights novel challenges and proposes mitigation strategies for securing ML systems.

This paper explores RL for cyber defense in SDN, resisting poisoning attacks.

problem Adversaries exploit ML adaptability to poison training and evade classification.
method Investigates RL algorithms for autonomous cyber defense in SDN, studying various attack types.
result RL agents can effectively react to poisoning attacks in SDN.

Deep learning enhances jamming and defense in wireless communications.

problem Improving wireless communication security against jamming attacks.
method Adversarial machine learning for jamming and a GAN for training data augmentation.
result Deep learning significantly reduces transmitter performance compared to random jamming.

New defense method for non-parametric classifiers robust against adversarial attacks.

problem Lack of robustness in non-parametric classifiers against adversarial attacks.
method Adversarial pruning method to preprocess datasets and a novel attack.
result Adversarial pruning provides a robust defense for non-parametric classifiers.

New defense mechanism detects and mitigates poisoned regression data.

problem Vulnerability of regression models to targeted data poisoning attacks.
method Introduces N-LID, a measure of local intrinsic dimensionality to distinguish poisoned samples.
result N-LID based defense outperforms state-of-the-art methods in prediction accuracy and runtime.

This paper discusses adversarial attacks on cyber security systems using machine learning.

problem Adversarial attacks limit the use of machine learning in cyber security.
method Characterizes adversarial attack methods and their applications in cyber security.
result Highlights unique challenges and future research directions for adversarial attacks in cyber security.