Detection and Localization of an Adversarial GPS Interference Source Based on Clock Signatures.
Abstract not provided.
Smith, Joseph↗
SEARCH · Engineering Papers
Results for “adversarial detection”
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.
Anomaly detection in Hyper Suprime-Cam galaxy images with generative adversarial networks
ABSTRACT The problem of anomaly detection in astronomical surveys is becoming increasingly important as data sets grow in size. We present the results of an unsupervised anomaly detection method using a Wasserstein generative adversarial network (WGAN) on nearly one million optical galaxy images in the Hyper Suprime-Cam (HSC) survey. The WGAN learns to generate realistic HSC-like galaxies that follow the distribution of the data set; anomalous images are defined based on a poor reconstruction by the generator and outlying features learned by the discriminator. We find that the discriminator is more attuned to potentially interesting anomalies compared to the generator, and compared to a simpler autoencoder-based anomaly detection approach, so we use the discriminator-selected images to construct a high-anomaly sample of ∼13 000 objects. We propose a new approach to further characterize these anomalous images: we use a convolutional autoencoder to reduce the dimensionality of the residual differences between the real and WGAN-reconstructed images and perform UMAP clustering on these. We report detected anomalies of interest including galaxy mergers, tidal features, and extreme star-forming galaxies. A follow-up spectroscopic analysis of one of these anomalies is detailed in the Appendix; we find that it is an unusual system most likely to be a metal-poor dwarf galaxy with an extremely blue, higher-metallicity H ii region. We have released a catalogue with the WGAN anomaly scores; the code and catalogue are available at https://github.com/kstoreyf/anomalies-GAN-HSC; and our interactive visualization tool for exploring the clustered data is at https://weirdgalaxi.es.
79 ASTRONOMY AND ASTROPHYSICS↗
Probing for Artifacts: Detecting Imagenet Model Evasions
While deep learning models have made incredible progress across a variety of machine learning tasks, they remain vulnerable to adversarial examples crafted to fool otherwise trustworthy models. In this work we approach this problem through the lens of a detection framework. We propose a classification network that uses the hidden layer activations of a trained model as inputs to detect adversarial artifacts in an input. We train this classification network simultaneously against multiple adversarial algorithms to create a more robust detector and show higher detection rates than several alternatives. The novelty of our approach is in the scale and scope of probing Imagenet models for adversarial artifacts. In addition, we propose an improvement to feature squeezing, another common adversarial example detection method.
Rounds, Jeremiah↗
An adversarial transformer for anomalous lamb wave pattern detection
Not Available
Guo, Jiawei (ORCID:0009000839628889)↗
Safe and Robust Binary Classification and Fault Detection Using Reinforcement Learning
In this paper, we propose a learning-based method utilizing the Soft Actor-Critic (SAC) algorithm to train a binary Support Vector Machine (SVM) classifier. This classifier is designed to identify valid input spaces in high-dimensional, highly constrained systems while minimizing the total runtime of offline simulations. The simulations adapt their runtime based on the likelihood that a given training input will be informative to the classifier. Furthermore, we introduce a method for using the trained SAC model to predict whether a desired system input is likely to violate constraints, along with a technique to adjust the input as necessary. Additionally, we explore the potential of this model to detect faults or adversarial attacks within the system. The effectiveness of our approach is demonstrated through various simulations of challenging classification problems and a constrained quadrotor model.
Netter, Josh [Georgia Institute of Technology, Atl↗
In-Device Delay: Introduction to Obscurant Physcial Protection System
The Office of Radiological Security (ORS) In-Device Delay (IDD) program has undertaken a project to research and develop a novel protection system for industrial irradiators that contain high-activity Co-60 sources. Based on adversary testing conducted by ORS, it is was determined that to successfully accomplish the theft of the target material, the adversary will require visual contact of the sources and source rack located at the bottom of the pool. Therefore, if a means of obscuring or visually hiding the sources in the pool can be achieved (while adhering to facility operations, safety, and regulatory requirements), then illicit source theft will be significantly hindered. This project aims to develop a low-cost, non-propriety obscurant that, when an adversary action is detected, the obscurant will be deployed into the pool quickly, rendering visual observation of the source problematic; however, this obscurant will not otherwise disturb the sources, source rack, and filtration system. The obscurant will remain in the pool until removed by another process.
61 RADIATION PROTECTION AND DOSIMETRY↗
Chess Master Project
This final technical report tracks the accomplishments of the Chess Master Project to the statement of project objects and resulting commercialization of the technology. Objectives for the project include 1) to sustain critical energy delivery functions during a cyber intrusion, control system operators need the ability to automate identification and containment of the affected network areas, and re-route critical information and control flows around; and 2) to effectively isolate impacted network areas and re-route critical flows, control system network operators need a global view of all the communication flows and have a method to proactively determine the whitelisted communications and how to respond to communications when adversarial behavior is detected.
42 ENGINEERING↗
Optimal helicopter trajectory planning for terrain following flight
Helicopters operating in high threat areas have to fly close to the earth surface to minimize the risk of being detected by the adversaries. Techniques are presented for low altitude helicopter trajectory planning. These methods are based on optimal control theory and appear to be implementable onboard in realtime. Second order necessary conditions are obtained to provide a criterion for finding the optimal trajectory when more than one extremal passes through a given point. A second trajectory planning method incorporating a quadratic performance index is also discussed. Trajectory planning problem is formulated as a differential game. The objective is to synthesize optimal trajectories in the presence of an actively maneuvering adversary. Numerical methods for obtaining solutions to these problems are outlined. As an alternative to numerical method, feedback linearizing transformations are combined with the linear quadratic game results to synthesize explicit nonlinear feedback strategies for helicopter pursuit-evasion. Some of the trajectories generated from this research are evaluated on a six-degree-of-freedom helicopter simulation incorporating an advanced autopilot. The optimal trajectory planning methods presented are also useful for autonomous land vehicle guidance.
Menon, P. K. A.↗
SafeDNN: Understanding and Verifying Neural Networks
The SafeDNN project at NASA Ames explores analysis techniques and tools to ensure that systems that use Deep Neural Networks (DNN) are safe, robust and interpretable. Research directions we are pursuing include: symbolic execution for DNN analysis, label-guided clustering to automatically identify input regions that are robust, parallel and compositional approaches to improve formal SMT-based verification, property inference and automated program repair for DNNs, adversarial training and detection, probabilistic reasoning for DNNs. In this talk I will highlight some of the research advances from SafeDNN, that were already published.
Corina Pasareanu↗
Systems and Implementation: Integrating Unmanned Aircraft Systems into Physical Protection Systems at Fixed Sites and During Transportation
Physical protection systems, and response forces in particular, are designed to prevent an adversary from successfully completing a malevolent act against a facility or transport operations. Timely detection and assessment of any potential adversary action against a target is an essential element of materials security. The timely detection and assessment must then be followed-up by a capable and timely response that might be enhanced with the additional situational awareness provided by unmanned aircraft systems (UAS). The United States Department of Energy’s National Nuclear Security Administration Office of International Nuclear Security has been exploring capabilities provided by UAS to support response force operations within the physical protection system. UAS have the potential to provide response force commanders and operators with situational awareness in assessing adversary locations and actions as well as the locations of responders. UAS may be utilized for area searches ahead of responder pathways to identify potential threats and to provide situational awareness of areas not normally covered by cameras (such as areas outside the fence line outside at fixed facilities). In addition, UAS can provide real-time information to transportation convoy teams that pass through constantly changing public access environments. This paper will provide operational recommendations to be addressed when integrating UAS into existing physical protection systems at fixed sites and during transport. Recommendations will include aspects of the following: needs analysis; tactics and techniques to support detection and assessment as well as response force deployment; remote pilot selection, qualifications, training, and currency; UAS selection criteria; UAS laws and regulations; possible cost sharing with other facility operations; and on-scene emergency management.
Stockwell, Brandon↗
Detecting CAN Masquerade Attacks with Signal Clustering Similarity
Vehicular Controller Area Networks (CANs) are susceptible to cyber attacks of different levels of sophistication. Fabrication attacks are the easiest to administer—an adversary simply sends (extra) frames on a CAN—but also the easiest to detect because they disrupt frame frequency. To overcome time-based detection methods, adversaries must administer masquerade attacks by sending frames in lieu of (and therefore at the expected time of) benign frames but with malicious payloads. Research efforts have proven that CAN attacks, and masquerade attacks in particular, can affect vehicle functionality. Examples include causing unintended acceleration, deactivation of vehicle’s brakes, as well as steering the vehicle. We hypothesize that masquerade attacks modify the nuanced correlations of CAN signal time series and how they cluster together. Therefore, changes in cluster assignments should indicate anomalous behavior. We confirm this hypothesis by leveraging our previously developed capability for reverse engineering CAN signals (i.e., CAN-D [Controller Area Network Decoder]) and focus on advancing the state of the art for detecting masquerade attacks by analyzing time series extracted from raw CAN frames. Specifically, we demonstrate that masquerade attacks can be detected by computing time series clustering similarity using hierarchical clustering on the vehicle’s CAN signals (time series) and comparing the clustering similarity across CAN captures with and without attacks. We test our approach in a previously collected CAN dataset with masquerade attacks (i.e., the ROAD dataset) and develop a forensic tool as a proof of concept to demonstrate the potential of the proposed approach for detecting CAN masquerade attacks.
Moriano Salazar, Pablo↗
Application of Quantum Machine Learning to High Energy Physics Analysis at LHC using IBM Quantum Computer Simulators and IBM Quantum Computer Hardware
Our group pioneers the use of Quantum Machine Learning (QML) on High Energy Physics analysis at LHC. We have successfully employed several QML classification algorithms in the ttH (Higgs production in association with a top quark pair) and Higgs to two muons (Higgs coupling to second generation fermions), two recent LHC flagship physics analysis, on gate-model quantum computer simulators and hardware. The simulation studies have been performed with the IBM Quantum Framework, Google Tensorflow Quantum Framework, and Amazon Braket Framework, and we have achieved good classification performance that is similar to the performances of the classical machine learning methods currently used in LHC physics analyses, classical SVM, classical BDT, and classical deep neural network for example. We have also performed our studies using IBM superconducting quantum computer hardware and the performance is promising and is approaching the performance from IBM quantum simulators. Moreover, we extend our studies to other QML areas such as quantum anomaly detection and quantum generative adversarial, and some preliminary results have been obtained. Also, we have overcome the challenges of intensive computing resources in the cases of large qubits (25 qubits or more) and large numbers of events using NVIDIA cuQuantum with NERSC Perlmutter HPC. Our studies give an example that Quantum Machine Learning performs as well as its classical counterpart for realistic High Energy Physics analysis datasets. Furthermore, our result on noisy quantum hardware provides important validation for the result on noiseless quantum simulators.
72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
Detecting False Data Injection Attacks in Smart Grids: A Semi-Supervised Deep Learning Approach
The dependence on advanced information and communication technology increases the vulnerability in smart grids under cyber-attacks. Recent research on unobservable false data injection attacks (FDIAs) reveals the high risk of secure system operation, since these attacks can bypass current bad data detection mechanisms. To mitigate this risk, this paper proposes a data-driven learning-based algorithm for detecting unobservable FDIAs in distribution systems. We use autoencoders for efficient dimension reduction and feature extraction of measurement datasets. Further, we integrate the autoencoders into an advanced generative adversarial network (GAN) framework, which successfully detects anomalies under FDIAs by capturing the unconformity between abnormal and secure measurements. Also, considering that the datasets collected from practical power systems are partially labeled due to expensive labeling costs and missing labels, the proposed method only requires a few labeled measurement data in addition to unlabeled data for training. Numerical simulations in three-phase unbalanced IEEE 13-bus and 123-bus distribution systems validate the detection accuracy and efficiency of this method.
97 MATHEMATICS AND COMPUTING↗
A Data Processing Pipeline for Adversarial Socio-Technical Network Analysis
With the rapid adoption of emerging technologies, there is a need to catalog and model sociotechnical interdependencies that have been historically used to influence the operation of Critical Infrastructure networks including the impacts of mergers and acquisitions, hostile takeovers, and foreign investment. Our research intends to address this need with two primary contributions. First, we have developed a data curation and processing pipeline to generate sociotechnical networks extracted from a variety of data sources including SEC filings and infrastructure asset databases. The pipeline, implemented in Apache Airflow, extracts and normalizes the representation of entities and relations, specified within ontologies. Our intent is to provide an extensible, machine-actionable approach to quickly communicate such models, reproduce previous results, and adapt them to new, unanticipated situations. Second, networks produced by our pipeline enable the development of graph-theoretic metrics that consider the properties of network components in addition to its topology. Metadata associated with network components---whether semantic, temporal, or geospatial---affects the alignment of generated networks with assumptions underlying complexity metrics. Validation of generated networks relative to component types defined by an ontology, may allow the research community to adapt metrics to the semantics of the domains being studied. Generated networks may be processed as knowledge, dynamic, or spatial graphs and enables a variety of analyses including automated reasoning and measures of network complexity. Automated reasoning views extracted entities and relations as a knowledge graph; this enables application of inference rules that represent historically-attested adversarial business methods and applies that behavior to a specific geographic context. Measures of network complexity, including degree distribution, reachability analyses, temporal analysis, and community detection can be adapted to indicate adversarial organizational influence.
97 MATHEMATICS AND COMPUTING↗
Utilization of the Critic Subnetwork of a Generative Adversarial Network as Detector of Morphological Material Change in Image Data
The resolution of computed tomography (CT) has become high enough to monitor morphological changes due to aging in materials in long-term applications. For this work, we explored the utility of the critic of a generative adversarial network (GAN) to automatically detect such changes. The GAN was trained with images of pristine Pharmatose, which is used as a surrogate energetic material. It is important to note that images of the material with altered morphology were only used during the test phase. The GAN-generated images reproduced the microstructure of Pharmatose well, although some unrealistic particle fusion was seen. Calculated morphological metrics (volume fraction, interfacial line length, and local thickness) for the synthetic images also showed good agreement with the training data, albeit with signs of mode collapse in the interfacial line length. While the critic exposed changes in particle size, it showed limited ability to distinguish images by particle shape. The detection of shape differences was also a more challenging task for the selected morphological metrics that related to energetic material performance. We further tested the critic with images of aged Pharmatose. Subtle changes due to aging are difficult for the human analyst to detect; but both critic and morphological metrics analysis showed image differentiation.
36 MATERIALS SCIENCE↗
Quantification of morphological change in materials based on image data utilizing machine learning techniques
Computed tomography (CT) resolution has become high enough to monitor morphological changes due to aging in materials in long-term applications. We explored the utility of the critic of a generative adversarial network (GAN) to automatically detect such changes. The GAN was trained with images of pristine Pharmatose, which is used as a surrogate energetic material. It is important to note that images of the material with altered morphology were only used during the test phase. The GAN-generated images visually reproduced the microstructure of Pharmatose well, although some unrealistic particle fusion was seen. Calculated morphological metrics (volume fraction, interfacial line length, and local thickness) for the synthetic images also showed good agreement with the training data, albeit with signs of mode collapse in the interfacial line length. While the critic exposed changes in particle size, it showed limited ability to distinguish images by particle shape. The detection of shape differences was also a more challenging task for the selected morphological metrics that related to energetic material performance. We further tested the critic with images of aged Pharmatose. Subtle changes due to aging are difficult for the human analyst to detect. Both critic and morphological metrics analysis showed image differentiation.
36 MATERIALS SCIENCE↗
Social Media Analytics Relevant to TikTok - a Literature Review and Directions for Future Research
We have attempted to capture a sense of the scientific state of the art in studying social media platforms, including data collection from platforms, understanding platform behavior, known adversarial uses, and adverse content detection, classification, and quantification. Our coverage of the field is backed up by roughly two hundred citations, and it concludes with a comparative analysis and a list of apparent gaps and potential paths forward.
99 GENERAL AND MISCELLANEOUS↗
Precursor Analysis Report: Industroyer2 and Wiper Malware Targeting Ukrainian Energy Provider 2022
The Industroyer2 and Wiper Malware Targeting Ukrainian Energy Provider 2022 Precursor Analysis Report leverages publicly available information about the Industroyer2 cyber attack and catalogs anomalous observables for each technique employed in the attack. This analysis is based upon the methodology of the Cybersecurity for the Operational Technology Environment (CyOTE) program. An adversary attempted to cause a blackout in Ukraine in April 2022 by using the Industroyer2 malware against a regional Ukrainian energy provider. The adversary targeted eight high-voltage electrical substations and utilized the malware in tandem with disk wipers for Windows, Linux, and Solaris operating systems in an attempt to make response and recovery efforts more difficult. The adversary reused a piece of the original Industroyer malware designed to open circuit breakers and de-energize target substations. The adversary gained initial access to the victim’s enterprise network through unknown means in February 2022 and was able to perform reconnaissance, pivot to the operations network, and reside in the system for at least 51 days. This gave the adversary a detailed understanding of the environment and allowed them to customize the Industroyer2 malware to the victim’s operations network. However, defenders detected and stopped the attack before the adversary could achieve their intended impact. Had the Industroyer2 attack been successful, it could have caused a blackout for more than two million people during the early stages of Russia’s invasion of Ukraine. Researchers and analysts identified 22 unique techniques (used in a sequence of 31 steps) utilized during the attack with a total of 297 observables using MITRE ATT&CK® for Industrial Control Systems. The CyOTE program assesses observables accompanying techniques used prior to the triggering event to identify opportunities to detect malicious activity. If observables accompanying the attack techniques are perceived and investigated prior to the triggering event, earlier comprehension of malicious activity can take place. Twenty-three of the identified techniques used during the Industroyer2 cyber attack were precursors to the triggering event. Analysis identified 224 observables associated with these precursor techniques, 122 of which were assessed to have an increased likelihood of being perceived in the 51 days preceding the triggering event. The response and comprehension time could have been reduced if the observables had been identified earlier. The information gathered in this report contributes to a library of observables tied to a repository of artifacts, data sources, and technique detection references for practitioners and developers to support the comprehension of indicators of attack. Asset owners and operators can use these products if they experience similar observables or to prepare for comparable scenarios.
45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗