SEARCH · Search NASA

Results for “adversarial attack”

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.

SourceResearch connectionFrom yearThrough year

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

5 records

All sources NASA NTRS DOE OSTI

NASA NTRSDate not supplied

NNREPAIR: Constraint-based Repair of Neural Network Classifiers

We present NNREPAIR, a constraint-based technique for repairing neural network classifiers by i) fixing the logic of the network at an intermediate layer, and ii) fixing the decision conditions of the network at the last layer. NNREPAIR first uses fault localization to find potentially faulty network parameters (specifically weights) and then performs repair using constraint solving to apply small modifications to remedy the defects. We present novel strategies to enable precise yet efficient repair such as inferring correctness specifications to act as oracles for intermediate layer repair, and generation of experts wrt each class. We demonstrate the technique in the context of three different scenarios: (1) Improving the over all accuracy of a model, (2) Fixing security vulnerabilities caused by poisoning of training data and (3) Improving the robustness of the network against adversarial attacks. Our evaluation on MNIST and CIFAR-10 models shows that NNREPAIR can improve the accuracy by 45.56% on poisoned data and 11.11% on adversarial data. NNREPAIR also provides improvement, albeit modest, in the overall accuracy of models, without requiring new data or re-training.

Repair↗

NASA NTRS2018

Engineering Trade-off Considerations Regarding Design-for-Security, Design-for-Verification, and Design-for-Test

The United States government has identified that application specific integrated circuit (ASIC) and field programmable gate array (FPGA) hardware are at risk from a variety of adversary attacks. This finding affects system security and trust. Consequently, processes are being developed for system mitigation and countermeasure application. The scope of this tutorial pertains to potential vulnerabilities and countermeasures within the ASIC/FPGA design cycle. The presentation demonstrates how design practices can affect the risk for the adversary to: change circuitry, steal intellectual property, and listen to data operations. An important portion of the design cycle is assuring the design is working as specified or as expected. This is accomplished by exhaustive testing of the target design. Alternatively, it has been shown that well established schemes for test coverage enhancement (design-for-verification (DFV) and design-for-test (DFT)) can create conduits for adversary accessibility. As a result, it is essential to perform a trade between robust test coverage versus reliable design implementation. The goal of this tutorial is to explain the evolution of design practices; review adversary accessibility points due to DFV and DFT circuitry insertion (back door circuitry); and to describe common engineering trade-off considerations for test versus adversary threats.

Design for reliability (DFR)↗

NASA NTRS2022

Neurosymbolic Hybrid Approach to Driver Collision Warning

There are two main algorithmic approaches to autonomous driving systems: (1) An end-to-end system in which a single deep neural network learns to map sensory input directly into appropriate warning and driving responses. (2) A mediated hybrid recognition system in which a system is created by combining independent modules that detect each semantic feature. While some researchers believe that deep learning can solve any problem, others believe that a more engineered and symbolic approach is needed to cope with complex environments with less data. Deep learning alone has achieved state-of-the-art results in many areas, from complex gameplay to predicting protein structures. In particular, in image classification and recognition, deep learning models have achieved accuracies as high as humans. But sometimes it can be very difficult to debug if the deep learning model doesn't work. Deep learning models can be vulnerable and are very sensitive to changes in data distribution. Generalization can be problematic. It's usually hard to prove why it works or doesn't. Deep learning models can also be vulnerable to adversarial attacks. Here, we combine deep learning-based object recognition and tracking with an adaptive neurosymbolic network agent, called the Non-Axiomatic Reasoning System (NARS), that can adapt to its environment by building concepts based on perceptual sequences. We achieved an improved intersection-over-union (IOU) object recognition performance of 0.65 in the adaptive retraining model compared to IOU 0.31 in the COCO data pre-trained model. We improved the object detection limits using RADAR sensors in a simulated environment, and demonstrated the weaving car detection capability by combining deep learning-based object detection and tracking with a neurosymbolic model.

Wang, Pei↗

NASA NTRS2015

Analyzing Cyber Security Threats on Cyber-Physical Systems Using Model-Based Systems Engineering

The spectre of cyber attacks on aerospace systems can no longer be ignored given that many of the components and vulnerabilities that have been successfully exploited by the adversary on other infrastructures are the same as those deployed and used within the aerospace environment. An important consideration with respect to the mission/safety critical infrastructure supporting space operations is that an appropriate defensive response to an attack invariably involves the need for high precision and accuracy, because an incorrect response can trigger unacceptable losses involving lives and/or significant financial damage. A highly precise defensive response, considering the typical complexity of aerospace environments, requires a detailed and well-founded understanding of the underlying system where the goal of the defensive response is to preserve critical mission objectives in the presence of adversarial activity. In this paper, a structured approach for modeling aerospace systems is described. The approach includes physical elements, network topology, software applications, system functions, and usage scenarios. We leverage Model-Based Systems Engineering methodology by utilizing the Object Management Group's Systems Modeling Language to represent the system being analyzed and also utilize model transformations to change relevant aspects of the model into specialized analyses. A novel visualization approach is utilized to visualize the entire model as a three-dimensional graph, allowing easier interaction with subject matter experts. The model provides a unifying structure for analyzing the impact of a particular attack or a particular type of attack. Two different example analysis types are demonstrated in this paper: a graph-based propagation analysis based on edge labels, and a graph-based propagation analysis based on node labels.

MBSE↗

NASA NTRS1972

Technology assessment or technology harassment: The attacks on science and technology

The manner in which technology is being assessed by various groups and individuals is discussed. Attacks on science and technology (specifically military uses and funding), and the disillusionment of the public with the lack of relevance of science to the public interest and with the infallible wisdom of scientists are described. The effects of the fear of environmental pollution are emphasized. It is concluded that the important issues of science, technology, and public policy will require a pluralistic definition of the public interest by open adversary proceedings. It is also felt that the danger is not that new technology will receive inadequate assessment for possible deleterious secondary effects, but that harassment by an overemotional political process may prevent its coming to fruition.

Green, L., Jr.↗