Slide decks, recorded presentations, posters, and related materials produced by Mississippi State University faculty, staff, and students.
-
Absence of Superconductivity in the lightly doped Hubbard model
Jodie Roberts and Rudolf Torsten Clay
The mechanism of superconductivity (SC) in high critical temperature cuprate superconductors remains an unsolved problem. The simplest electronic model of cuprate superconductors is the one band Hubbard model. It models copper atom positions with lattice sites, omitting oxygen atoms for simplicity, and separates the Hamiltonian into hopping (t,t) and interaction (U) components. The simplest Hubbard model only considers nearest neighbor hopping, t. To account for overlap between oxygen orbitals also requires next nearest neighbor hopping, t'. Exact solutions of the model are computationally prohibitive to find for large systems. Quantum Monte Carlo (QMC) methods such as Constrained Path Monte Carlo (CPMC) can be used in cases where exact methods cannot. Constraining the imaginary time path removes the Fermion sign problem caused by sign degeneracy of Slater determinants. However, an additional approximate technique known as back propagation must also be used to measure any quantity besides the energy within CPMC. A newly proposed released constraint measurement method instead releases the path constraint of CPMC for short intervals. This is more accurate but reintroduces the sign problem. We present the first calculations of superconducting pair-pair correlations in the Hubbard model using the released constraint technique. Our results show that in general back propagation tends to underestimate long- range superconducting pairing in the Hubbard model. Recent work using CPMC has suggested that SC does exist in the lightly doped two-dimensional Hubbard model. Our results show that superconducting pair-pair correlations continuously weaken with increasing U, suggesting that SC is not present.
-
Anomaly Detection in Material Images Using Scan Statistics
Kaitlyn Anderson, Asanka Duwage, and Tung-Lung Wu
Nanoproducts are a growing sector due to their unique properties and wide range of applications across industries. However, nanomaterial production is a complex process that requires a high degree of precision, making it challenging to ensure consistent quality at a large scale. Minor defects can significantly alter their functional properties and overall performance, making accurate detection of defects crucial for maintaining and controlling nanomaterial properties. To address these challenges, this project applies scan statistics to detect localized defects in scanning election microscope images of nanofibrous materials. We implemented both square and circular scanning windows of varying sizes to identify clusters in the images. By generating null distributions through permutation testing, we assessed the statistical significance of detected regions, leading to reliable identification of anomalies. For each image, we identified the most effective scan window by selecting the size that minimized the p-value, allowing us to adapt the detection process to the unique spatial features of each image and effectively address different types of anomalies. Overall, this approach provides a robust and adaptable method for automated anomaly detection, with the potential to enhance quality control in nanomaterial manufacturing.
-
Computational Investigation of Substituent Effects on Zirconium Pincer Complexes via Density Functional Theory Methods
Anna Constable, Garrett M. Wells, Samuel D. Juarez Escamilla, Thedford K. Hollis, and Charles Edwin Webster
Organic light-emitting diodes (OLEDs) have emerged as a promising technology for displays due to their high efficiency and superior color performance. The primary objective of this study was to investigate the tunability of zirconium carbene "pincer" complexes for potential use in OLED applications. Specifically, we aim to understand how changes in coordinated ligands and their substituents influence the electronic structure and spectral properties. A series of computational tests were conducted using density functional theory (DFT) to optimize the ground-state geometries and time-dependent density functional theory (TD-DFT) to optimize the excited-state geometries. TD-DFT calculations are also used to predict absorption and emission spectra. The effects of ligand variation and the simulated spectra will be discussed.
-
Developing a Computational Pipeline for Microstructure-based Modelling with ExaCA and EVPFFT
Lizzy Beall, Eric Collins, and Jacob Moore
The microstructure of a metal determines its properties and by understanding the grains that make up that structure, we can predict the behavior of that material. However, it can be difficult and costly to view the microstructure of a metal, especially since the microstructure is highly dependent on the manufacturing history of the part. By computer generating the microstructure of a material, we can better understand its properties. Exascale Cellular Automata (ExaCA) can generate a microstructure for a metal sample given its thermal history and Elasto-Visco Plastic Fast Fourier Transforms (EVPFFT) can model the response of the crystals in a grain structure to deformation. The focus of this investigation is the development of a work method to use both of these programs, compare their fidelity to physical reality and create one script to automatically run both programs in sequence. First, the structure needs to be generated in ExaCA, which for simplicity is a 128x128x128 voxelized cube directionally cooled in the Z direction. The output is converted into an input file for EVPFFT using a python conversion script. Then EVPFFT is used to model that microstructure generated by ExaCA under different conditions, in this case, tension in the Z direction. Running both of these programs from one script can increase the efficiency of the process. This modelling pipeline can be used to generate high fidelity data to increase the accuracy of predictions based on the grain structure of a material and train reduced order models.
-
Developing a Neural Network for Prediction of Interatomic Energies in Iron-Manganese Alloys
Robert D. H. Race, Doyl Dickel, and Hala Ben Messaoud
Machine-learned interatomic potentials (ML IAP) for alloy systems have been sought after due to their ability to reduce experimentation costs and time, accelerating alloy development and discovery. However, the explicit inclusion of magnetism in these potentials has been both a difficult and important problem to solve, due to the complexity of spin-lattice dynamics and its significance in the properties of magnetic alloys. We present here the development of an explicitly magnetic Fe-Mn ML IAP using a physics informed neural network (PINN) extension of the rapid artificial neural network (RANN) formalism. It is shown that the potential is capable of reproducing a number of energetic, mechanical, and magnetic properties of the Fe-Mn system, including phase stability and magnetic ordering, as well as thermal and elastic properties. The presented formalism and potential should provide a useful platform for the exploration of magnetic alloy systems and their properties at the atomistic scale.
-
Efficient Image Denoising Models with Anderson Acceleration using Finite Difference Methods
Amanda E. Diegel, Spence Hanegan, Hyeona Lim, and Hoang Tran
Current image denoising algorithms based on variational methods can suffer from slow convergence or no convergence due to high nonlinearity of the images. To speed up the convergence of denoising, we apply Anderson acceleration to the fixed-point image denoising problem. Anderson acceleration is an algorithmic method for reducing the number of fixed-point iterations necessary for convergence. It involves using weighted updates to each iteration based on the weighted residuals from past iterations, or history. By using finite difference methods, we can approximate the gradient and higher-order partial derivatives at points on the image. We then use these approximations to create matrix equations to solve for a denoised image. By iterating and applying Anderson Acceleration, we achieve a faster convergence of image denoising. This method is tested and compared to the fixed-point method and other conventional denoising methods using peak signal to noise ratio (PSNR).
-
Finite Element Methods with Anderson Acceleration and its Application to Image Denoising
Amanda E. Diegel, Spence Hanegan, Hyeona Lim, and Hoang Tran
Image denoising is an important computational tool with applications in the medical, material science, and defense fields where CT-scans have a lot of noise that degrades quality and clearness. While there are several methods of solving image denoising problems, the one we focused on is total variation where we solve a difficult nonlinear partial differential equation that minimizes noise. There are also many numerical methods to find an approximate solution to this nonlinear partial differential equation, but the one we focus on is the finite element method. In addition, we used a fixed-point iteration method to handle the nonlinearity and obtain convergence. But the main problem arises in the number of iterations needed to achieve convergence due to the complexity of the nonlinear equations. So, we propose implementing Anderson Acceleration to speed up the fixed-point iteration method. In addition, we propose adding length and angle filtering to Anderson Acceleration to reduce redundant data and get convergence quicker. We used MATLAB along with FELICTY: Finite Element Implementation and Computational Interface Tool for You toolbox to execute the computations.
-
Investigating the Defect Behavior and Electronic Properties of Formamidinium Lead Bromide Perovskite through Machine Learned Interatomic Potentials
Rachel Lee, John Michael Lane, Woodrow Wilson, and Neeraj Rai
Formamidinium lead bromide (FAPbBr3) perovskite crystals display promising optoelectronic properties, making them attractive materials for solar cells, light-emitting diodes, and photoelectrochemical devices. In this study, computational methods are applied to investigate the effects of point defects—vacancies, interstitials, Frenkel, and Schottky—on the electronic structure of FAPbBr3 bulk and nanoplatelet crystals. Traditional computational methods, such as density functional theory (DFT), are computationally expensive and thus limited to small systems and short time scales. To characterize and simulate larger systems over longer time periods at reduced computational costs, machine learned interatomic potentials (MLIPs) are developed with MACE, a message passing neural network. The machine learning model is trained on ab initio molecular dynamics (AIMD) simulations of defect-free and defectpresent systems to run molecular dynamics (MD) simulations. Several models are trained using different combinations of AIMD systems data, with the model trained on defect-free bulk FAPbBr3 systems being the most accurate. The MACE MD simulations provide key insights into the change in band structure and density of states of FAPbBr3 crystals with defects present. This study illustrates the effectiveness of MLIPs in capturing defect-driven behavior in FAPDBr3, enabling accurate, large-scale MD simulations and informing the more efficient use of FAPbBr3 and intelligent design of perovskites in optoelectronic applications.
-
Neural Network Prediction of Titanium-Aluminum Mechanical Properties
Carter Reed and Kip Barrett
Technological progress is strongly coupled with our understanding of materials. In fields like aerospace, titanium-aluminum (Ti-Al) alloys are of particular interest for their strength-to-weight ratio and high-temperature performance. The ability to accurately predict these characteristics as a function of composition and processing conditions would reduce research costs by tuning the performance to the application's needs before producing the physical material. The mechanical properties, and performance by proxy, of a material are dependent on the microstructure of and the interatomic interactions within the material. Thus, we validate the rapid artificial neural network (RANN) potential of the titanium-aluminum binary system for the prediction of mechanical properties via tensile stress simulations in the LAMMPS software package. The RAN potential has predicted the simulation yield strength for a perfect crystal of pure titanium to be 15 GPa. We see that by randomly making about 5% of the hcp sites monovacancies, the yield strength was reduced to approximately half of the perfect crystal, while a spherical void with a radius of 5 lattice parameters had a similar but less pronounced effect. It was also found that by applying a grain boundary perpendicular to the basal plane of one grain and parallel to the basal plane of the other grain, the yield strength was reduced to about a fourth of that of the perfect crystal. By visualizing the simulations in OVITO, we observed the formation and disappearance of bcc layers as well as both deformation twinning and recrystallization twinning.
-
Virtual screening of peptides that can prevent insulin aggregation
Thanh Tien Dao, Bidisha Sengupta, and Steven Gwaltney
Diabetes is a growing health concern, with almost 3% of the population of the United States using insulin injections to control blood sugar levels. Insulin is prone to aggregation during storage and injection. The toxic products of aggregation can cause an increase in the required dosage to achieve the desired therapeutic effect. The driving intermediate of aggregation is believed to be a partially folded insulin, derived from the insulin monomer. We hypothesize that stabilizing the insulin monomer with a peptide may prevent this unfolding process and subsequent aggregation. However, the space of all possible peptides is impossibly large to study systematically. Therefore, we have generated a set of 2,000 randomly chosen 20-mer peptides. We have generated 3D structures of the proposed peptide sequence using AlphaFold 2 and have utilized the peptide-protein docking software Autodock CrankPep (ADCP) to determine how each peptide binds to the insulin monomer. Our results show that different peptides have drastically different binding geometries and binding energies. The next step will be to use the results of the docking studies to train a neural network that can predict the binding of any peptide sequence to the insulin monomer. Additionally, molecular dynamics simulations will be run on promising sequences, to identify short peptides that can prevent insulin aggregation.
-
A docking Analysis of reactivators for sarin-inhibited acetylcholinestrase can aid in screening candidate compounds
Ryan Pirger and Steven Gwaltney
Poster created as part of the Center for Computational Sciences' Research Experiences for Undergraduates in Computational Methods with Applications in Materials Science and presented at the 2024 Undergraduate Research Showcase.
Organophosphate (OP) poisoning disrupts nerve signaling by inhibiting acetylcholinesterase (AChE), an essential enzyme. Current oxime-based treatments for OP poisoning lack efficacy in severe cases due to blood-brain barrier (BBB) impermeability. A prior machine learning study identified promising AChE reactivators with BBB permeability and synthetic feasibility. Our study employed in silico docking simulations using AutoDock to evaluate the interactions between these potential reactivators and a sarin-inhibited human AChE model. The 35 compounds proposed in the earlier study, along with five known good AChE reactivators, were docked against a model of human AChE that was inhibited by the OP nerve agent sarin. While all the positive control molecules yielded good docking results, none of the newly proposed compounds did. The results of this docking analysis will inform the development of novel OP antidotes capable of reaching the brain and effectively reactivating AChE in sever poisoning scenarios. We suggest adding a docking screening to any future protocol designed to generate potential reactivators.
-
Alkylidyne Molybdenum formation from Lithiated Transmetalation: A Computational Insight
Niles Wahlin, Nghia Le, and Charles Edwin Webster
Alkylidyne complexes have garnered significant interest for advances in catalysis in various industrial processes, such as Fischer-Tropsch synthesis and alkyne metathesis. Recently, the alkylidyne complex [N3N]MoCH was synthesized by reacting [N3N]MoCI with lithiated methanoanthracene. Despite the structural similarity between norbornadiene and methanoanthracene, as well as the overall exergonic nature of the reactions, the reaction ony proceeds experimental with lithiated methanoanthracene. To shed light on this transformation, which involves multiple spin states of various short-lived intermediates, we report a computational study elucidating the structure-reactivity relationship of lithiated cyclic reagents. The reaction proceeds through a transmetallation step followed by an endergonic carbon transfer. The carbon transfer is expected to occur via a stepwise mechanism, breaking two C-C bonds separately. The short-lived intermediate after transmetallation can potentially be observed through absorption spectroscopy; thus, simulated spectra and Natural Transition Orbital (NTO) analysis are performed for reference. The proposed mechanism from the computational results suggests a switch in spin states from triplet to singlet during the reaction. We extend our study by examining the reaction between a molybdenum complex wth a pyridine-based PNP pincer ligand (PNP = 2,5-bis(di-tert-butylphosphinomethyl)pyrrolide) and lithiated methanoanthracene.
-
Continuous Data Assimilation for Two-Phase Flow
Mary H. Graveman and Amanda E. Diegel
Poster created as part of the Center for Computational Sciences' Research Experiences for Undergraduates in Computational Methods with Applications in Materials Science and presented at the 2024 Undergraduate Research Showcase.
We propose a numerical approximation method modeling two-phase flow (via the Cahn-Hilliard equation) that incorporates data to achieve long-time accuracy. The underlying numerical method utilizes the Galerkin finite element method for spatial discretization and a method known as continuous data assimilation to incorporate the known data. We demonstrate the method is long-time stable and long-time accurate provided enough data measurements are incorporated into the simulation, overcoming possibly inaccurate initial conditions. Numerical experiments illustrate the effectiveness of the method on a benchmark test problem. All computations are completed in MATLAB.
-
Developing Machine Learned Interatomic Potentials for Hydrogen Dissociation over Molybdenum Phosphide
Jeremy Lugo, John Michael Lane, Woodrow N. Wilson, and Neeraj Rai
Poster created as part of the Center for Computational Sciences' Research Experiences for Undergraduates in Computational Methods with Applications in Materials Science and presented at the 2024 Undergraduate Research Showcase.
Molybdenum phosphides are promising catalysts for biomass conversion, particularly in the hydrogenation of organic molecules. Hydrogen dissociation on the surface of such catalysts is a critical step in the process. Traditional computational methods such as density functional theory (DFT) can provide valuable insights into hydrogen-catalyst interactions but are limited by their high computational cost and inefficiency for large systems and long time scales. Machine learning has emerged as a solution to these limitations. Machine learned interatomic potentials (MLIPs), trained on ab initio molecular dynamics (AIMD) simulations, enable the development of molecular dynamics (MD) for much larger systems over much longer time scales, with significantly reduced computational cost. This allows more complex interactions and rare reactive events to be observed that would otherwise be inaccessible with traditional methods. In this study, MACE, a message passing neural network, is employed to develop MLIPs for the dissociation of hydrogen on molybdenum phosphide surfaces. Utilizing MACE results in highly accurate MD simulations with time scales on the order of 1,000 times longer than those achievable with AIMD. These MD simulations provide key insights about the system, including preferred adsorption sites and atomic interactions and serve as a bridge to studying biomass conversion. This study demonstrates the efficacy of machine learning for catalytic research, illustrating the robust and efficient predictive capabilities of MLIPs and underscoring the potential of machine learning techniques for the efficient discovery, optimization, and understanding of new catalysts.
-
Development of a Simulation Tool to Study Secondary Electron Emission in Copper Coated with Graphene
George Vassilakopoulos and Eric Collins
Poster created as part of the Center for Computational Sciences' Research Experiences for Undergraduates in Computational Methods with Applications in Materials Science and presented at the 2024 Undergraduate Research Showcase.
Secondary electron emission (SEE) arises from the interaction of high-energy particles with metallic surface materials. The emission of electrons from copper surfaces can be problematic in a variety of applications, such as particle accelerators or microchips operating in spacecraft. Emitted electrons can build up in accelerator walls, reducing the accuracy of the experiment. SEE in microchips can flip logical bits, create unwanted heat, and reduce the device's lifespan. In recent years, graphene has been proposed as a potential moderator to prevent or mitigate these effects. Graphene can be applied in layers on top of a copper substrate to reduce the secondary electron yield (SEY): the ratio of emitted secondary electrons to total incident electrons. The focus of this investigation has been to simulate the SEE process in copper and to obtain its SEY at various energies of incident primary electrons. Monte Carlo methods, such as the continuous-slowing-down approximation and numerical integration schemes, have been used to generate and track secondary electrons within the copper. The simulation generates secondary electrons from primary electron interactions consistent with existing material models. The positions and energy losses of the secondary electrons are updated using Mott's cross-section formula and the stopping power, respectively. Electrons’ paths are then propagated until their energy falls below a prescribed threshold or the electron is emitted from the surface. Simulations have successfully reproduced SEY results for copper with and without graphene surface layers for an incident electron energy range of 60 eV to 1000 eV. The results show that the SEY of the copper with graphene layers is reduced by 20% compared to the SEY of pure copper.
-
Drawing a Positive Mathematics Identity: Portrait of a Maths Person
Liza Bondurant and Liesl McConchie
In this poster we’ll share a powerful classroom activity we designed that aims to broaden students’ definition of what a maths person looks like. We implemented this activity with Black and Latinx middle school students. We noticed a significant difference in students’ portraits with a simple intervention, exposing students to mathematicians from cultures who have historically been marginalized from maths spaces, specifically Black and Latinx mathematicians. These findings suggest that being introduced to mathematicians of color can help broaden students’ perceptions of who belongs in maths spaces and support students’ development of the confidence and skills they need to succeed in maths class and beyond.
-
Embodied Noticing in Mathematics: Pre-service Teachers’ Observations and Noticings
Jonathan Troup, Liza Bondurant, Claudia Bertolone-Smith, Diana Moss, and Hortensia Soto
-
Engaging Preservice Elementary Teachers in Statistical Investigations of Systemic Racism in School Discipline Data
Liza Bondurant, Anthony Fernandes, Ksenija Simić-Muller, and Travis Weiland
In this poster, we share preservice elementary teachers' (PSTs') initial noticings during exploratory data analysis of materials focused on developing content knowledge of statistics as well as normalizing conversations of race in math class. The content covered is typically included in college-level introduction to statistics courses. We summarize PSTs' noticings and wonderings of an interactive data dashboard presenting racial disparities in school discipline.
-
New Weighting Parameters for Non-Local Means based Denoising Algorithm
Ely Carroll and Hyeona Lim
Poster created as part of the Center for Computational Sciences' Research Experiences for Undergraduates in Computational Methods with Applications in Materials Science and presented at the 2024 Undergraduate Research Showcase.
Current image denoising techniques can be broadly categorized into two types: variational methods which use minimizing functionals, and filtering methods via pixel averaging. Specifically, the non-local means (NLM) filtering based image denoising algorithm involves replacing each pixel value by the weighted averages of all pixels in the entire image. The basic non-local means algorithm is predicated on manually inputting a parameter to determine the weight of each pixel. We introduce and analyze a new method of implementing the weight factors, based on small variances in the pixel values to identify noise. The new method will identify newly classified noisy pixels, then while comparing blocks of pixels for similarity it is possible to disregard noisy pixels and gain a clearer comparison of structure for weighting. The new method is numerically tested and compared to conventional NLM based methods via the peak signal to noise ratio (PSNR), and visual comparison. The results show that the new method is preferable to the current NLM methods.
-
Repulsive Coulomb interactions enhance Superconductivity Selectively at Density 0.5 per site
Jeremy Padvorac and Rudolf Torsten Clay
Poster created as part of the Center for Computational Sciences' Research Experiences for Undergraduates in Computational Methods with Applications in Materials Science and presented at the 2024 Undergraduate Research Showcase.
The globally accepted Bardeen-Cooper-Schrieffer (BCS) theory explains the pairing mechanism in low-temperature elemental superconductors but fails to explain pairing in high critical temperature superconductors. Unconventional superconductors with higher critical temperatures require moreexotic pairing mechanisms considering electron-electron interactions, such as spin-fluctuated-mediated pairing. Many-body calculations within the Hubbard model, which includes short-range electron-electron repulsion, have shown that superconducting pair correlations are enhanced when the density of carriers is close to one-half per orbital, a characteristic density in many unconventional superconductors. However, questions related to the accuracy of previous calculations remain. In this work, we employed a more accurate numerical method, Self-Consistent Constrained Path Quantum Monte Carlo (SC-CPMC), to check the accuracy of CPMC calculations without the self-consistent optimization of the trial wavefunction. By eliminating the artificial lattice symmetry breaking previously used, we restored the intrinsic symmetry of the lattice. This study also investigated effects of a larger Hubbard U. Our results confirm that the Hubbard U significantly enhancessuperconducting pair-pair correlations for carrier densities close to 0.5 per site.
-
Using an LMS to Restructure ETD Submission: A Five-Year Review
Teri M. Robinson and Danny D. Davis
At Mississippi State University, theses and dissertations are reviewed and approved by the Office of Thesis and Dissertation Format Review (OTD). OTD is a part of the University’s Library and works closely with the Graduate School to set submission standards. In 2019 OTD began using Canvas, the University’s learning management system (LMS), for all thesis and dissertation submissions. Students are expected to enroll in a required course during their final semester to gain access to the submission modules in Canvas. All forms and documents are submitted in Canvas, making the process entirely electronic. While the Canvas course has proved to be easy for students to navigate, it has not been everything the OTD expected or hoped for when creating the course. There are administrative tasks that were not anticipated, reporting tools not working as intended, and difficulty getting all of campus to understand and come on board with the new procedures. This presentation will cover how the process has evolved over the last five years using Canvas and cover some of the benefits and disadvantages discovered along the way.
-
Validation and Runtime Improvements in Neural Networks for Interatomic Potentials via Automatic Fingerprint Selection
Spencer Evans-Cole, Kip Barrett, and Doyl Dickel
Poster created as part of the Center for Computational Sciences' Research Experiences for Undergraduates in Computational Methods with Applications in Materials Science and presented at the 2024 Undergraduate Research Showcase.
The development of interatomic potentials via machine learning on first principles results is driven by the need to accurately and quickly analyze atomic environments. When encoding atomic environments for neural networks, a series of different two and three-body functions are used to transcribe the local environments to produce a vector of inputs, making it imperative that the local environments are encoded optimally. Given a set of atomic environments, we generate a series of different fingerprints and by analysis of the data select the most optimal ones which demonstrate a reduced validation error and computation time in our model. To select the best fingerprints for the network, we first construct a large dense matrix A, where each row is an atomic environment and each column is a fingerprint with a different set of parameters. We look to create the most optimal rank k approximation of A, by factoring it as A = CUR where C is made up of k columns of A, R is made up of k rows of A and U is square and of rank no greater than k. We employ the singular value decomposition followed by two rounds of sampling the column space of our matrix, where we utilize weights based on spectral properties and then adaptively compute leverage scores to select fingerprints that were not initially chosen and that require a low computational cost with high probability. By computing the matrices C and R we find the optimal representation of our fingerprint and training data respectively. Our optimal fingerprints achieve on average a 39% reduction in validation error with comparable training errors to naive fingerprint selections for unary and binary systems, while minimizing the computational cost for fingerprint generation.
-
Comprehensive Wind Speed Forecasting-Based Analysis of Stacked Stateful & Stateless Models
Swayamjit Saha, Amogu Uduka, Hunter Walt, and James Lucore
Wind speed is a powerful source of renewable energy, which can be used as an alternative to the non-renewable resources for production of electricity. Renewable sources are clean, infinite and do not impact the environment negatively during production of electrical energy. However, while eliciting electrical energy from renewable resources viz. solar irradiance, wind speed, hydro should require special planning failing which may result in huge loss of labour and money for setting up the system. In this poster, we discuss four deep recurrent neural networks viz. Stacked Stateless LSTM, Stacked Stateless GRU, Stacked Stateful LSTM and Statcked Stateful GRU which will be used to predict wind speed on a short-term basis for the airport sites beside two campuses of Mississippi State University. The paper does a comprehensive analysis of the performance of the models used describing their architectures and how efficiently they elicit the results with the help of RMSE values. A detailed description of the time and space complexities of the above models has also been discussed.
-
Examining Curriculum Requirements of Undergraduate Teacher Preparation Programs to Gauge Educator Knowledge of Trauma-Informed Education Frameworks
Laura Grace King
Statistics show that many pre-K–12 students in the United States are exposed to adverse childhood experiences (ACEs) or potentially traumatic events (PTEs), and a growing body of literature points to these childhood experiences harming children’s academic functioning and future educational attainment. A review of the literature on trauma-informed education (i.e., curriculum and programs designed to mitigate the negative effects of trauma) highlights many teachers’ lack of confidence in combatting issues within student populations affected by adversity and trauma; the research also indicated that teachers with knowledge of trauma/adversity and its implications are crucial to effectively educating at-risk children and adolescents. The current study examines teacher education curricula at 119 postsecondary institutions accredited by both the Southern Association of Colleges and Schools Commission on Colleges (SACSCOC) and the Council for the Accreditation of Educator Preparation (CAEP) to identify existing course requirements that fit each of four dimensions of many trauma-informed education programs: Adversity and Resilience, Human/Child Development, Child/Educational Psychology, and Human/Cultural Diversity. Primary and secondary education majors were examined separately for inclusion of these dimensions and compared. Results show that only one of the 119 universities’ primary education programs required courses in Adversity and Resilience; none were required among the secondary education programs. At least one course in Human/Child Development was required by 52% of primary education programs and 46% of secondary education programs; Child/Educational Psychology by 55% and 51%, respectively; and Human/Cultural Diversity by 57% and 50%, respectively. This examination forms part of an ongoing evaluation of teacher education standards and educator preparedness to implement trauma-informed education interventions. Results suggest a need for additional training and professional development for educators, especially given the prevalence of child traumatic stress and the growing number of policies and initiatives promoting trauma-sensitive schools.
-
The Impact of Single vs. Poly Victimization by Maltreatment Type on Pre-Treatment PTSD Scores
Ashley G. O'Donnell, Sabrina DiCarlo, and Arazais D. Oliveros
Studies show that an estimated 21-50% of children who have experienced maltreatment will develop post-traumatic stress disorder (PTSD) within their lifetime (Schuck & Widom 2019). Research suggests that the type and number of exposures to trauma may influence symptom severity. Further, trauma symptom scores of children who experienced child sexual abuse (CSA) were higher among the children who endorsed poly-victimization, meaning when CSA was combined with another form of maltreatment (Racine et al 2022). This study seeks to examine the impact of poly-victimization and combined types of maltreatment on children’s PTSD scores in an archival dataset from a child advocacy center (CAC) serving children exposed to various forms of trauma. Participants from the overall child sample (N = 721) who have pre-treatment PTSD scores (n = 290) will be analyzed. The analyzed sample includes 83 minors exposed to poly-victimization, 175 exposed to single victimization, and 32 where the trauma type was missing from the dataset. Specifically, the following hypotheses will be tested: (1) victims of poly-victimization will have higher PTSD scores than victims of single victimization; (2) among participants with poly-victimization, those with a combination of sexual abuse and physical abuse will have the highest PTSD scores compared to other combinations. Results will be discussed in the context of current referral pathways for child advocacy centers.
