Search NASA⌕ Search

SEARCH · Search NASA

Results for “Adversarial Machine Learning”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 records

Deep Deception: Exemplars of Adversarial Machine Learning and Countermeasures Applicable to International Safeguards

As a follow-up to our more comprehensive report on Adversarial Machine Learning (AML), here we provide demonstrations of AML attacks against the Limbo image database of UF6 cylinders in a variety of orientations and amongst a variety of distractor images. We demonstrate the Carlini & Wagner AML attack against a subset of Limbo images, with 100% attack success rate; meaning all attacked images were misclassified by a highly accurate trained model, yet the image changes were imperceptible to the human eye. We also demonstrate successful attacks against segmented images (images with more than one targeted object). Finally, we demonstrated the Fast Fourier Transform countermeasure that can be used to detect AML attacks on images. The intent of this and our previous report is to inform the IAEA and stakeholders of both the promise of machine learning, which could greatly improve the efficiency of surveillance monitoring, but also of the real threat of AML and potential defenses.

97 MATHEMATICS AND COMPUTING↗

Attack on Grid Event Cause Analysis: An Adversarial Machine Learning Approach

With the ever-increasing reliance on data for data-driven applications in power grids, such as event cause analysis, the authenticity of data streams has become crucially important. The data can be prone to adversarial stealthy attacks aiming to manipulate the data such that residual-based bad data detectors cannot detect them, and the perception of system operators or event classifiers changes about the actual event. This paper investigates the impact of adversarial attacks on convolutional neural network-based event cause analysis frameworks. We have successfully verified the ability of adversaries to maliciously misclassify events through stealthy data manipulations. The vulnerability assessment is studied with respect to the number of compromised measurements. Furthermore, a defense mechanism to robustify the performance of the event cause analysis is proposed. The effectiveness of adversarial attacks on changing the output of the framework is studied using the data generated by real-time digital simulator (RTDS) under different scenarios such as type of attacks and level of access to data.

Niazazari, Iman↗

3D reconstruction and neural rendering for adversarial machine learning

While evasion attacks on computer vision systems have been widely studied, creating attacks that remain effective under significant changes in viewpoint continues to be challenging. Traditional approaches often rely on affine transformations of images, but these approaches degrade at larger perspective shifts and often produce unrealistic or ineffective perturbations. Recent methods use differentiable renderers to improve viewpoint robustness, but they typically depend on manually constructed 3D models. We introduce a semi-automated pipeline that generates physically printable and perspective-invariant adversarial patches using only a small set of 2D images. Our method integrates 3D reconstruction, neural rendering, adversarial patch optimization, and an object detection victim model into a unified workflow. We use 2D Gaussian Splatting for high fidelity mesh reconstruction and FlexPara for surface parameterization that produces texture maps suitable for patch editing. Together, these components form a fully differentiable pipeline in PyTorch3D that links texture modification to model outputs, enabling efficient optimization of patches that remain effective across many viewpoints. The complete process, from image capture to patch printing and physical evaluation, can be completed within a few hours. We demonstrate the effectiveness of the resulting patches through attacks on the YOLOv8 object detection model and discuss remaining challenges and opportunities for improving robustness and scalability.

Singhvi, Vivaan [ORNL] (ORCID:0009000586288221)↗

Performance Evaluation of Vertical Federated Machine Learning Against Adversarial Threats on Wide-Area Control System: Preprint

Federated machine learning (FL) is gaining significant popularity to develop cybersecurity solutions in power grids because of its advanced capability to support decentralized data handing at local devices, its privacy preservation, and its low-bandwidth requirement. However, the evolving adversarial machine learning (AML) threats raise significant concerns for the cybersecurity of FL architectures. The FL-based split neural network (SplitNN) achieves high performance through the decentralized training of local neural network models while preserving data privacy across multiple entities. In this paper, we propose a methodology for evaluating the performance of a vertical FLbased anomaly detector against different types of AML attacks, including denial-of-service attacks, adversarial data injection attacks, and replay attacks on the trained local models deployed in the grid network. For a case study, we consider the modified IEEE 13-bus system, and we develop SplitNN-based binary and multiclass classification models to detect, locate, and identify different types of data integrity attacks on the volt-watt control with two pooling layers: maximum pooling and AvgPool. Our experimental results, computed through performance metrics, reveal that the severity of these AML attacks varies with the integrated pooling mechanism, the type of classification model, and the nature of the cyberattack. Further, the AML attacks negatively impacted the prediction time per sample for the pretrained SplitNN during the online testing.

adversarial threats↗

Using Machine Learning in Adversarial Environments.

Intrusion/anomaly detection systems are among the first lines of cyber defense. Commonly, they either use signatures or machine learning (ML) to identify threats, but fail to account for sophisticated attackers trying to circumvent them. We propose to embed machine learning within a game theoretic framework that performs adversarial modeling, develops methods for optimizing operational response based on ML, and integrates the resulting optimization codebase into the existing ML infrastructure developed by the Hybrid LDRD. Our approach addresses three key shortcomings of ML in adversarial settings: 1) resulting classifiers are typically deterministic and, therefore, easy to reverse engineer; 2) ML approaches only address the prediction problem, but do not prescribe how one should operationalize predictions, nor account for operational costs and constraints; and 3) ML approaches do not model attackers’ response and can be circumvented by sophisticated adversaries. The principal novelty of our approach is to construct an optimization framework that blends ML, operational considerations, and a model predicting attackers reaction, with the goal of computing optimal moving target defense. One important challenge is to construct a realistic model of an adversary that is tractable, yet realistic. We aim to advance the science of attacker modeling by considering game-theoretic methods, and by engaging experimental subjects with red teaming experience in trying to actively circumvent an intrusion detection system, and learning a predictive model of such circumvention activities. In addition, we will generate metrics to test that a particular model of an adversary is consistent with available data.

97 MATHEMATICS AND COMPUTING↗

Using Machine Learning in Adversarial Environments

Cyber defense is an asymmetric battle today. We need to understand better what options are available for providing defenders with possible advantages. Our project combines machine learning, optimization, and game theory to obscure our defensive posture from the information the adversaries are able to observe. The main conceptual contribution of this research is to separate the problem of prediction, for which machine learning is used, and the problem of computing optimal operational decisions based on such predictions, coupled with a model of adversarial response. This research includes modeling of the attacker and defender, formulation of useful optimization models for studying adversarial interactions, and user studies to measure the impact of the modeling approaches in realistic settings.

97 MATHEMATICS AND COMPUTING↗

Defending Against Adversarial Examples

Adversarial machine learning is an active field of research that seeks to investigate the security of machine learning methods against cyber-attacks. An important branch of this field is adversarial examples, which seek to trick machine learning models into misclassifying inputs by maliciously tampering with input data. As a result of the pervasiveness of machine learning models in diverse areas such as computer vision, health care, and national security, this vulnerability is a rapidly growing threat. With the increasing use of AI solutions, threats against AI must be considered before deploying systems in a contested space. Adversarial machine learning is a problem strongly tied to software security, and just like other more common software vulnerabilities, it exploits a weakness in software, like components of machine learning models. During this project, we attempted to survey and replicate several adversarial machine learning techniques with the goal of developing capabilities for Sandia to advise and defend against these threats. To accomplish this, we scanned state of the art research for robust defenses against adversarial examples and applied them to a machine learning problem.

97 MATHEMATICS AND COMPUTING↗

Robust Power System Stability Assessment Against Adversarial Machine Learning-Based Cyberattacks via Online Purification

The increasing complexity associated with renewable generation brings more challenges to power system stability assessment (SA). Data-driven approaches based on machine learning (ML) techniques for stability assessment have received significant research interest and shown their promising performance. However, ML-based models are recognized to be vulnerable to adversarial disturbances, where a slight perturbation to power system measurements could lead to unacceptable errors. To address this issue, this paper develops a novel lightweight mitigation strategy, i.e., robust online stability assessment (ROSA), to enhance the ML-based assessment model against both white-box and the black-box adversarial disturbances (i.e., purification) in the online implementation. The ROSA involves a supervised learning-based module for the primary stability assessment and a self-supervised learning-based module. Further, the two modules are trained jointly with different objective (loss) functions and implemented in sequence. A suitable purification objective and various time-series data augmentation methods are designed for SA applications to tackle adversarial disturbances adaptively. Case studies are performed, and the comparative results have clearly illustrated the competitive, robust accuracy against various adversarial scenarios and verified the effectiveness of the proposed online purification strategy.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Adversarial Sampling-Based Motion Planning

In this report there are many scenarios in which a mobile agent may not want its path to be predictable. Examples include preserving privacy or confusing an adversary. However, this desire for deception can conflict with the need for a low path cost. Optimal plans such as those produced by RRT* may have low path cost, but their optimality makes them predictable. Similarly, a deceptive path that features numerous zig-zags may take too long to reach the goal. We address this trade-off by drawing inspiration from adversarial machine learning. We propose a new planning algorithm, which we title Adversarial RRT*. Adversarial RRT* attempts to deceive machine learning classifiers by incorporating a predicted measure of deception into the planner cost function. Adversarial RRT* considers both path cost and a measure of predicted deceptiveness in order to produce a trajectory with low path cost that still has deceptive properties. We demonstrate the performance of Adversarial RRT*, with two measures of deception, using a simulated Dubins vehicle. We show how Adversarial RRT* can decrease cumulative RNN accuracy across paths to 10%, compared to 46% cumulative accuracy on near-optimal RRT* paths, while keeping path length within 16% of optimal. We also present an example demonstration where the Adversarial RRT* planner attempts to safely deliver a high value package while an adversary observes the path and tries to intercept the package.

42 ENGINEERING↗

Introduction to the special section on learning and security for multi-agent systems

This report features six articles that provide an overview of the emerging research topics at the intersection learning, security, and multi-agent systems. Recent years have witnessed a surge in the number of works at their intersections, and they have appeared in system and control communities as well as many other communities in artificial intelligence, cyber–physical systems, and economics. The articles in this special section give accessible and comprehensive tutorials and surveys for a broad systems and control audience, covering topics including adversarial machine learning, multi-agent reinforcement learning, cyber resilience, resilient control systems, and game design. It is hopeful that this special section will spawn future interest and cross-disciplinary collaborations in this emerging transdisciplinary research area.

97 MATHEMATICS AND COMPUTING↗

Assembling a Cyber Range to Evaluate Artificial Intelligence / Machine Learning (AI/ML) Security Tools

In this case study, we will describe the design and assembly of a cyber security test range we have built at Oak Ridge National Laboratory in Oak Ridge, TN, USA. The range is designed to provide a flexible environment to evaluate cyber security tools—particularly those involving AI/ML—in a way that provides realistic environments and where we can control the experiments to determine the strengths and weaknesses of the tools. We have designed in the ability to repeat the evaluations, so additional tools can be evaluated and compared at a later time. The system is one that can be scaled up or down for experiment sizes. At the time of the conference we will have completed two full-scale, national, government challenges on this range. These challenges are evaluating the performance and operating costs for AI/ML-based cyber security tools for application into large, government-sized environments. These evaluations will be described, in order to provide motivation and context for various design decisions and adaptations we have made. The first challenge measured end-point security tools against 100K malware samples chosen across a range of types. The second is network detection of attempted penetration and exploitations with varying levels of covertness in a high-volume, business network. The scale of each of these challenges is requiring us to create automation systems to repeat the experiments identically for each tool. Preventing there being easy signs of malicious activity for the AI/ML tools to focus on has been a particularly interesting and challenging aspect of designing and executing these challenge events. After the events, the range continues to be used for other research such as adversarial machine learning where the repeatability, scale, and automation required for the national challenge events become essential elements for research.

Nichols, Jeff↗