-
DRL-GAN: A Hybrid Approach for Binary and Multiclass Network Intrusion Detection
Authors:
Caroline Strickland,
Chandrika Saha,
Muhammad Zakar,
Sareh Nejad,
Noshin Tasnim,
Daniel Lizotte,
Anwar Haque
Abstract:
Our increasingly connected world continues to face an ever-growing amount of network-based attacks. Intrusion detection systems (IDS) are an essential security technology for detecting these attacks. Although numerous machine learning-based IDS have been proposed for the detection of malicious network traffic, the majority have difficulty properly detecting and classifying the more uncommon attack…
▽ More
Our increasingly connected world continues to face an ever-growing amount of network-based attacks. Intrusion detection systems (IDS) are an essential security technology for detecting these attacks. Although numerous machine learning-based IDS have been proposed for the detection of malicious network traffic, the majority have difficulty properly detecting and classifying the more uncommon attack types. In this paper, we implement a novel hybrid technique using synthetic data produced by a Generative Adversarial Network (GAN) to use as input for training a Deep Reinforcement Learning (DRL) model. Our GAN model is trained with the NSL-KDD dataset for four attack categories as well as normal network flow. Ultimately, our findings demonstrate that training the DRL on specific synthetic datasets can result in better performance in correctly classifying minority classes over training on the true imbalanced dataset.
△ Less
Submitted 5 January, 2023;
originally announced January 2023.
-
Agent-based Dynamics of a SPAHR Opioid Model on Social Network Structures
Authors:
Owen Queen,
Vincent Jodoin,
Leigh B. Pearcy,
W. Christopher Strickland
Abstract:
Addiction epidemiology has been an active area of mathematical research in recent years. However, the social and mental processes involved in substance use disorders versus contraction of a pathogenic disease have presented challenges to advancing the epidemiological theory of substance abuse, especially within the context of the opioids where both prescriptions and social contagion have played a…
▽ More
Addiction epidemiology has been an active area of mathematical research in recent years. However, the social and mental processes involved in substance use disorders versus contraction of a pathogenic disease have presented challenges to advancing the epidemiological theory of substance abuse, especially within the context of the opioids where both prescriptions and social contagion have played a major role. In this paper, we utilize an agent-based modeling approach on social networks to further explore these dynamics. Using parameter estimation approaches, we compare our results to that of the Phillips et al. SPAHR model which was previously fit to data from the state of Tennessee. Our results show that the average path length of a social network has a strong relationship to social contagion dynamics for drug use initiation, while other pathways to substance use disorder should not be constrained to social network interactions that predate the individual's drug use.
△ Less
Submitted 15 February, 2022;
originally announced February 2022.
-
Transverse Hydraulic Fracture Initiation: Insights from Intermediate-Scale Hydraulic Fracture Tests
Authors:
Jeffrey Burghardt,
Hunter Knox,
Chris Strickland,
Vincent Vermeul,
Pengcheng Fu,
Craig Ulrich,
Mark McClure,
Mathew Ingraham,
Paul Schwering,
Timothy Kneafsey,
Doug Blankenship
Abstract:
In a uniform stress field, a tensile hydraulic fracture will favor propagation in the plane formed by the maximum (most compressive) and intermediate principal stresses. For the propagation of hydraulic fractures in non-uniform stress fields, especially those where the principal stress directions change spatially, things become much more complex, are less well understood, and therefore are more di…
▽ More
In a uniform stress field, a tensile hydraulic fracture will favor propagation in the plane formed by the maximum (most compressive) and intermediate principal stresses. For the propagation of hydraulic fractures in non-uniform stress fields, especially those where the principal stress directions change spatially, things become much more complex, are less well understood, and therefore are more difficult to predict. Such is often the case when a hydraulic fracture is initiated from a borehole drilled in the direction of the minimum principal stress. This is a well-known problem that has been the subject of several numerical and experimental studies. The purpose of this paper is to offer new insights from a series of hydraulic fracture tests conducted in sub-horizontal boreholes drilled in the Sanford Underground Research Facility. This project, called the Enhanced Geothermal Systems (EGS) Collab project, has developed a test bed at a depth of approximately 1478 meters underground consisting of sub-horizontal injection, production, and monitoring wells. This paper presents stimulation test data, and supporting geologic/geophysical characterization data. The test data are compared to previous numerical and experimental studies to determine how well this new experimental data compares with previous modeling, laboratory, and field tests.
△ Less
Submitted 1 October, 2021;
originally announced October 2021.
-
Agent-based and continuous models of hopper bands for the Australian plague locust: How resource consumption mediates pulse formation and geometry
Authors:
Andrew J. Bernoff,
Michael Culshaw-Maurer,
Rebecca A. Everett,
Maryann E. Hohn,
W. Christopher Strickland,
Jasper Weinburd
Abstract:
Locusts are significant agricultural pests. Under favorable environmental conditions flightless juveniles may aggregate into coherent, aligned swarms referred to as hopper bands. These bands are often observed as a propagating wave having a dense front with rapidly decreasing density in the wake. A tantalizing and common observation is that these fronts slow and steepen in the presence of green ve…
▽ More
Locusts are significant agricultural pests. Under favorable environmental conditions flightless juveniles may aggregate into coherent, aligned swarms referred to as hopper bands. These bands are often observed as a propagating wave having a dense front with rapidly decreasing density in the wake. A tantalizing and common observation is that these fronts slow and steepen in the presence of green vegetation. This suggests the collective motion of the band is mediated by resource consumption. Our goal is to model and quantify this effect. We focus on the Australian plague locust, for which excellent field and experimental data is available. Exploiting the alignment of locusts in hopper bands, we concentrate solely on the density variation perpendicular to the front. We develop two models in tandem; an agent-based model that tracks the position of individuals and a partial differential equation model that describes locust density. In both these models, locust are either stationary (and feeding) or moving. Resources decrease with feeding. The rate at which locusts transition between moving and stationary (and vice versa) is enhanced (diminished) by resource abundance. This effect proves essential to the formation, shape, and speed of locust hopper bands in our models. From the biological literature we estimate ranges for the ten input parameters of our models. Sobol sensitivity analysis yields insight into how the band's collective characteristics vary with changes in the input parameters. By examining 4.4 million parameter combinations, we identify biologically consistent parameters that reproduce field observations. We thus demonstrate that resource-dependent behavior can explain the density distribution observed in locust hopper bands. This work suggests that feeding behaviors should be an intrinsic part of future modeling efforts.
△ Less
Submitted 24 February, 2020; v1 submitted 31 October, 2019;
originally announced October 2019.
-
Classification of Protein Crystallization X-Ray Images Using Major Convolutional Neural Network Architectures
Authors:
Soheil Ghafurian,
Peter Orth,
Corey Strickland,
Hua Su,
Sangita Patel,
Steven Soisson,
Belma Dogdas
Abstract:
The generation of protein crystals is necessary for the study of protein molecular function and structure. This is done empirically by processing large numbers of crystallization trials and inspecting them regularly in search of those with forming crystals. To avoid missing the hard-gained crystals, this visual inspection of the trial X-ray images is done manually as opposed to the existing less a…
▽ More
The generation of protein crystals is necessary for the study of protein molecular function and structure. This is done empirically by processing large numbers of crystallization trials and inspecting them regularly in search of those with forming crystals. To avoid missing the hard-gained crystals, this visual inspection of the trial X-ray images is done manually as opposed to the existing less accurate machine learning methods. To achieve higher accuracy for automation, we applied some of the most successful convolutional neural networks (ResNet, Inception, VGG, and AlexNet) for 10-way classification of the X-ray images. We showed that substantial classification accuracy is gained by using such networks compared to two simpler ones previously proposed for this purpose. The best accuracy was obtained from ResNet (81.43%), which corresponds to a missed crystal rate of 5.9%. This rate could be lowered to less than 0.1% by using a top-3 classification strategy. Our dataset consisted of 486,000 internally annotated images, which was augmented to more than a million to address class imbalance. We also provide a label-wise analysis of the results, identifying the main sources of error and inaccuracy.
△ Less
Submitted 11 May, 2018;
originally announced May 2018.
-
Modeling the prescription opioid epidemic
Authors:
Nicholas A. Battista,
Leigh B. Pearcy,
W. Christopher Strickland
Abstract:
Opioid addiction has become a global epidemic and a national health crisis in recent years, with the number of opioid overdose fatalities steadily increasing since the 1990s. In contrast to the dynamics of a typical illicit drug or disease epidemic, opioid addiction has its roots in legal, prescription medication - a fact which greatly increases the exposed population and provides additional drug…
▽ More
Opioid addiction has become a global epidemic and a national health crisis in recent years, with the number of opioid overdose fatalities steadily increasing since the 1990s. In contrast to the dynamics of a typical illicit drug or disease epidemic, opioid addiction has its roots in legal, prescription medication - a fact which greatly increases the exposed population and provides additional drug accessibility for addicts. In this paper, we present a mathematical model for prescription drug addiction and treatment with parameters and validation based on data from the opioid epidemic. Key dynamics considered include addiction through prescription, addiction from illicit sources, and treatment. Through mathematical analysis, we show that no addiction-free equilibrium can exist without stringent control over how opioids are administered and prescribed, effectively transforming the dynamics of the opioid epidemic into those found in a purely illicit drug model. Numerical sensitivity analysis suggests that relatively low states of endemic addiction can be obtained by primarily focusing on medical prevention followed by aggressive treatment of remaining cases - even when the probability of relapse from treatment remains high. Further empirical study focused on understanding the rate of illicit drug dependence versus overdose risk, along with the current and changing rates of opioid prescription and treatment, would shed significant light on optimal control efforts and feasible outcomes for this epidemic and drug epidemics in general.
△ Less
Submitted 2 June, 2018; v1 submitted 9 November, 2017;
originally announced November 2017.
-
Three-dimensional low Reynolds number flows near biological filtering and protective layers
Authors:
W. Christopher Strickland,
Laura A. Miller,
Arvind Santhanakrishnan,
Christina Hamlet,
Nicholas A. Battista,
Virginia Pasour
Abstract:
Mesoscale filtering and protective layers are replete throughout the natural world. Within the body, arrays of extracellular proteins, microvilli, and cilia can act as both protective layers and mechanosensors. For example, blood flow profiles through the endothelial surface layer determine the amount of shear stress felt by the endothelial cells and may alter the rates at which molecules enter an…
▽ More
Mesoscale filtering and protective layers are replete throughout the natural world. Within the body, arrays of extracellular proteins, microvilli, and cilia can act as both protective layers and mechanosensors. For example, blood flow profiles through the endothelial surface layer determine the amount of shear stress felt by the endothelial cells and may alter the rates at which molecules enter and exit the cells. Characterizing the flow profiles through such layers is therefore critical towards understanding the function of such arrays in cell signaling and molecular filtering. External filtering layers are also important to many animals and plants. Trichomes (the hairs or fine outgrowths on plants) can drastically alter both the average wind speed and profile near the leaf's surface, affecting the rates of nutrient and heat exchange. In this paper, dynamically scaled physical models are used to study the flow profiles outside of arrays of cylinders that represent such filtering and protective layers. In addition, numerical simulations using the Immersed Boundary Method are used to resolve the 3D flows within the layers. The experimental and computational results are compared to analytical results obtained by modeling the layer as a homogeneous porous medium with free flow above the layer. The experimental results show that the bulk flow is well described by simple analytical models. The numerical results show that the spatially averaged flow within the layer is well described by the Brinkman model. The numerical results also demonstrate that the flow can be highly 3D with fluid moving into and out of the layer. These effects are not described by the Brinkman model and may be significant for biologically relevant volume fractions. The results of this paper can be used to understand how variations in density and height of such structures can alter shear stresses and bulk flows.
△ Less
Submitted 14 September, 2017;
originally announced September 2017.
-
IB2d Reloaded: a more powerful Python and MATLAB implementation of the immersed boundary method
Authors:
Nicholas Battista,
Christopher Strickland,
Aaron Barrett,
Laura Miller
Abstract:
The immersed boundary method (IB) is an elegant way to fully couple the motion of a fluid and deformations of an immersed elastic structure. In that vein, the IB2d software allows for expedited explorations of fluid-structure interaction for beginners and veterans to the field of computational fluid dynamics (CFD). While most open source CFD codes are written in low level programming environments,…
▽ More
The immersed boundary method (IB) is an elegant way to fully couple the motion of a fluid and deformations of an immersed elastic structure. In that vein, the IB2d software allows for expedited explorations of fluid-structure interaction for beginners and veterans to the field of computational fluid dynamics (CFD). While most open source CFD codes are written in low level programming environments, IB2d was specifically written in high- level programming environments to make its accessibility extend beyond scientists with vast programming experience. Although introduced previously by Battista et al. 2015, many improvements and additions have been made to the software to allow for even more robust models of material properties for the elastic structures, including a data analysis package for both the fluid and immersed structure data, an improved time-step** scheme for higher accuracy solutions, and functionality for modeling slight fluid density variations as given by the Boussinesq approximation.
△ Less
Submitted 21 July, 2017;
originally announced July 2017.
-
IB2d: a Python and MATLAB implementation of the immersed boundary method
Authors:
Nicholas A. Battista,
W. Christopher Strickland,
Laura A. Miller
Abstract:
The development of fluid-structure interaction (FSI) software involves trade-offs between ease of use, generality, performance, and cost. Typically there are large learning curves when using low-level software to model the interaction of an elastic structure immersed in a uniform density fluid. Many existing codes are not publicly available, and the commercial software that exists usually requires…
▽ More
The development of fluid-structure interaction (FSI) software involves trade-offs between ease of use, generality, performance, and cost. Typically there are large learning curves when using low-level software to model the interaction of an elastic structure immersed in a uniform density fluid. Many existing codes are not publicly available, and the commercial software that exists usually requires expensive licenses and may not be as robust or allow the necessary flexibility that in house codes can provide. We present an open source immersed boundary software package, IB2d, with full implementations in both MATLAB and Python, that is capable of running a vast range of biomechanics models and is accessible to scientists who have experience in high-level programming environments. IB2d contains multiple options for constructing material properties of the fiber structure, as well as the advection-diffusion of a chemical gradient, muscle mechanics models, and artificial forcing to drive boundaries with a preferred motion.
△ Less
Submitted 24 October, 2016;
originally announced October 2016.
-
Annealed Important Sampling for Models with Latent Variables
Authors:
M. -N. Tran,
C. Strickland,
M. K. Pitt,
R. Kohn
Abstract:
This paper is concerned with Bayesian inference when the likelihood is analytically intractable but can be unbiasedly estimated. We propose an annealed importance sampling procedure for estimating expectations with respect to the posterior. The proposed algorithm is useful in cases where finding a good proposal density is challenging, and when estimates of the marginal likelihood are required. The…
▽ More
This paper is concerned with Bayesian inference when the likelihood is analytically intractable but can be unbiasedly estimated. We propose an annealed importance sampling procedure for estimating expectations with respect to the posterior. The proposed algorithm is useful in cases where finding a good proposal density is challenging, and when estimates of the marginal likelihood are required. The effect of likelihood estimation is investigated, and the results provide guidelines on how to set up the precision of the likelihood estimation in order to optimally implement the procedure. The methodological results are empirically demonstrated in several simulated and real data examples.
△ Less
Submitted 24 February, 2014;
originally announced February 2014.
-
Scalable iterative methods for sampling from massive Gaussian random vectors
Authors:
Daniel P. Simpson,
Ian W. Turner,
Christopher M. Strickland,
Anthony N. Pettitt
Abstract:
Sampling from Gaussian Markov random fields (GMRFs), that is multivariate Gaussian ran- dom vectors that are parameterised by the inverse of their covariance matrix, is a fundamental problem in computational statistics. In this paper, we show how we can exploit arbitrarily accu- rate approximations to a GMRF to speed up Krylov subspace sampling methods. We also show that these methods can be used…
▽ More
Sampling from Gaussian Markov random fields (GMRFs), that is multivariate Gaussian ran- dom vectors that are parameterised by the inverse of their covariance matrix, is a fundamental problem in computational statistics. In this paper, we show how we can exploit arbitrarily accu- rate approximations to a GMRF to speed up Krylov subspace sampling methods. We also show that these methods can be used when computing the normalising constant of a large multivariate Gaussian distribution, which is needed for both any likelihood-based inference method. The method we derive is also applicable to other structured Gaussian random vectors and, in particu- lar, we show that when the precision matrix is a perturbation of a (block) circulant matrix, it is still possible to derive O(n log n) sampling schemes.
△ Less
Submitted 5 December, 2013;
originally announced December 2013.
-
A Bayesian changepoint methodology for high dimensional multivariate time series and space-time data: A study of structural change using remotely sensed data
Authors:
Chris Strickland,
Robert Burdett,
Robert Denham,
Robert Kohn,
Kerrie Mengersen
Abstract:
A Bayesian approach is developed to analyze change points in multivariate time series and space-time data. The methodology is used to assess the impact of extended inundation on the ecosystem of the Gulf Plains bioregion in northern Australia. The proposed approach can be implemented for dynamic mixture models that have a conditionally Gaussian state space representation. Details are given on how…
▽ More
A Bayesian approach is developed to analyze change points in multivariate time series and space-time data. The methodology is used to assess the impact of extended inundation on the ecosystem of the Gulf Plains bioregion in northern Australia. The proposed approach can be implemented for dynamic mixture models that have a conditionally Gaussian state space representation. Details are given on how to efficiently implement the algorithm for a general class of multivariate time series and space-time models. This efficient implementation makes it feasible to analyze high dimensional, but of realistic size, space-time data sets because our approach can be appreciably faster, possibly millions of times, than a standard implementation in such cases.
△ Less
Submitted 19 June, 2013;
originally announced June 2013.
