An official website of the United States government
We study rare-event simulation for a class of problems where the target hitting sets of interest are defined via modern machine learning tools such as neural networks and random forests. This problem is motivated from fast emerging studies on the safety evaluation of intelligent systems, robustness quantification of learning models, and other potential applications to large-scale simulation in which machine learning tools can be used to approximate complex rare-event set boundaries. We investigate an importance sampling scheme that integrates the dominating point machinery in large deviations and sequential mixed integer programming to locate the underlying dominating points. Our approach works for a range of neural network architectures including fully connected layers, rectified linear units, normalization, pooling and convolutional layers, and random forests built from standard decision trees. We provide efficiency guarantees and numerical demonstration of our approach using a classification model in the UCI Machine Learning Repository.
more »
« less
Estimation of cyber network risk using rare event simulation
Inherent vulnerabilities in a cyber network’s constituent machine services can be exploited by malicious agents. As a result, the machines on any network are at risk. Security specialists seek to mitigate the risk of intrusion events through network reconfiguration and defense. When dealing with rare cyber events, high-quality risk estimates using standard simulation approaches may be unattainable, or have significant attached uncertainty, even with a large computational simulation budget. To address this issue, an efficient rare event simulation modeling and analysis technique, namely, importance sampling for cyber networks, is developed. The importance sampling method parametrically amplifies certain aspects of the network in order to cause a rare event to happen more frequently. Output collected under these amplified conditions is then scaled back into the context of the original network to provide meaningful statistical inferences. The importance sampling methodology is tailored to cyber network attacks and takes the attacker’s successes and failures as well as the attacker’s targeting choices into account. The methodology is shown to produce estimates of higher quality than standard simulation with greater computational efficiency.
more »
« less
- Award ID(s):
- 1526383
- PAR ID:
- 10190302
- Date Published:
- Journal Name:
- The Journal of Defense Modeling and Simulation: Applications, Methodology, Technology
- ISSN:
- 1548-5129
- Page Range / eLocation ID:
- 154851292093455
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
An Efficient Multifidelity Model for Assessing Risk Probabilities in Power Systems under Rare EventsRisk assessment of power system failures induced by low-frequency, high-impact rare events is of paramount importance to power system planners and operators. In this paper, we develop a cost-effective multi-surrogate method based on multifidelity model for assessing risks in probabilistic power-flow analysis under rare events. Specifically, multiple polynomial-chaos-expansion-based surrogate models are constructed to reproduce power system responses to the stochastic changes of the load and the random occurrence of component outages. These surrogates then propagate a large number of samples at negligible computation cost and thus efficiently screen out the samples associated with high-risk rare events. The results generated by the surrogates, however, may be biased for the samples located in the low-probability tail regions that are critical to power system risk assessment. To resolve this issue, the original high-fidelity power system model is adopted to fine-tune the estimation results of low-fidelity surrogates by reevaluating only a small portion of the samples. This multifidelity model approach greatly improves the computational efficiency of the traditional Monte Carlo method used in computing the risk-event probabilities under rare events without sacrificing computational accuracy.more » « less
-
Arbey, Alexandre; Bélanger, G.; Desai, Nishita; Gonzalo, Tomas; Harlander, Robert V. (Ed.)A trio of automated collider event analysis tools are described and demonstrated, in the form of a quick-start tutorial. AEACuS interfaces with the standard MadGraph/MadEvent, Pythia, and Delphes simulation chain, via the Root file output. An extensive algorithm library facilitates the computation of standard collider event variables and the transformation of object groups (including jet clustering and substructure analysis). Arbitrary user-defined variables and external function calls are also supported. An efficient mechanism is provided for sorting events into channels with distinct features. RHADAManTHUS generates publication-quality one- and two-dimensional histograms from event statistics computed by AEACuS, calling MatPlotLib on the back end. Large batches of simulation (representing either distinct final states and/or oversampling of a common phase space) are merged internally, and per-event weights are handled consistently throughout. Arbitrary bin-wise functional transformations are readily specified, e.g. for visualizing signal-to-background significance as a function of cut threshold. MInOS implements machine learning on computed event statistics with XGBoost. Ensemble training against distinct background components may be combined to generate composite classifications with enhanced discrimination. ROC curves, as well as score distribution, feature importance, and significance plots are generated on the fly. Each of these tools is controlled via instructions supplied in a reusable cardfile, employing a simple, compact, and powerful meta-language syntax.more » « less
-
Caltais, Georgiana; Schilling, Christian (Ed.)Rare events are known to potentially cause pathological behavior in biochemical reaction systems. It is important to understand the cause. However, rare events are challenging to analyze due to their extremely low observability. This paper presents a fully automated approach that rapidly generates a large number of execution traces guaranteed to reach user-specified rare-event states for Chemical Reaction Network (CRN) models. It is enabled by a unique combination of a multi-layered and service-oriented CRN formal modeling approach, a dependency graph method to aid the shortest rare-event trace generation, and randomized compositional testing. The resulting prototype tool shows marked improvement over stochastic simulation and probabilistic model checking and it offers insights into a CRN.more » « less
-
Artificial intelligence has recently been widely used in computational imaging. The deep neural network (DNN) improves the signal-to-noise ratio of the retrieved images, whose quality is otherwise corrupted due to the low sampling ratio or noisy environments. This work proposes a new computational imaging scheme based on the sequence transduction mechanism with the transformer network. The simulation database assists the network in achieving signal translation ability. The experimental single-pixel detector’s signal will be ‘translated’ into a 2D image in an end-to-end manner. High-quality images with no background noise can be retrieved at a sampling ratio as low as 2%. The illumination patterns can be either well-designed speckle patterns for sub-Nyquist imaging or random speckle patterns. Moreover, our method is robust to noise interference. This translation mechanism opens a new direction for DNN-assisted ghost imaging and can be used in various computational imaging scenarios.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1177/1548512920934551
