Recent Integrations of Latent Variable Network Modeling With Psychometric Models

Wang, Selena

doi:10.3389/fpsyg.2021.773289

Cited by 5 publications

(3 citation statements)

References 76 publications

(95 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Latent variables have been used to simplify learning processes or create interpretable models in many fields; here, we use them primarily to simplify learning when we have multiple target variables. ,− We hypothesized that replacing a complex, all-encompassing, and opaque model with an ensemble of simpler models would greatly simplify the entire learning process based on the physical insights noted above. More generally, the right combination of weak learners has been proven to be quite effective, as shown by boosted decision trees .…”

Section: Resultsmentioning

confidence: 99%

Latent Variable Machine Learning Framework for Catalysis: General Models, Transfer Learning, and Interpretability

Kayode,

Montemore

2023

JACS Au

View full text Add to dashboard Cite

Machine learning has been successfully applied in recent years to screen materials for a variety of applications. However, despite recent advances, most screening-based machine learning approaches are limited in generality and transferability, requiring new models to be created from scratch for each new application. This is particularly apparent in catalysis, where there are many possible intermediates and transition states of interest in addition to a large number of potential catalytic materials. In this work, we developed a new machine learning framework that is built on chemical principles and allows the creation of general, interpretable, reusable models. Our new architecture uses latent variables to create a set of submodels that each take on a relatively simple learning task, leading to higher data efficiency and promoting transfer learning. This architecture infuses fundamental chemical principles, such as the existence of elements as discrete entities. We show that this architecture allows for the creation of models that can be reused for many different applications, providing significant improvements in efficiency and convenience. For example, our architecture allows simultaneous prediction of adsorption energies for many adsorbates on a broad array of alloy surfaces with mean absolute errors (MAEs) around 0.20−0.25 eV. The integration of latent variables provides physical interpretability, as predictions can be explained in terms of the learned chemical environment as represented by the latent space. Further, these latent variables also serve as new feature representations, allowing efficient transfer learning. For example, new models with useful levels of accuracy can be created with less than 10 data points, including transfer learning to an experimental data set with an MAE less than 0.15 eV. Lastly, we show that our new machine learning architecture is general and robust enough to handle heterogeneous and multifidelity data sets, allowing researchers to leverage existing data sets to speed up screening using their own computational setup.

show abstract

Section: Resultsmentioning

confidence: 99%

Latent Variable Machine Learning Framework for Catalysis: General Models, Transfer Learning, and Interpretability

Kayode,

Montemore

2023

JACS Au

View full text Add to dashboard Cite

show abstract

“…A Análise de Rede tem sido Revista Iberoamericana de Diagnóstico y Evaluacióne Avaliação Psicológica. RIDEP • Nº68 • Vol.2 • 165-180 • 2023 cada vez mais utilizada como uma análise complementar para a validação de instrumentos (Andrade et al, 2022;Wang, 2021). Os grafos foram gerados utilizando-se o método Least Absolute Shrinkage and Selection Operator (LASSO), que estima as correlações entre os itens da escala usando correlações parciais.…”

Section: Estrutura Fatorialunclassified

Propriedades Psicométricas da Fear of COVID-19 Scale e Avaliação de Medo em Adolescentes Brasileiros

Oliveira¹,

Enumo²,

Silva³

et al. 2023

REV IBEROAM DIAGN EV

View full text Add to dashboard Cite

Esse estudo objetivou avaliar as propriedades psicométricas da Fear of COVID-19 Scale em uma amostra de adolescentes brasileiros, comparando-as por sexo e orientação sexual. Foi realizado um estudo de natureza exploratória do qual participaram 176 adolescentes (M=15,5 anos). Foram aplicados remotamente um questionário de caracterização, a Fear of COVID-19 Scale e a Depression, Anxiety and Stress Scale 21. Os dados foram descritos e desenvolvidas: análise de variância, análise fatorial confirmatória, análise de rede e teste de validade convergente e preditiva. Os dados revelaram que a escala de medo pode ser aplicada na população adolescente, sendo capaz de avaliar o medo relacionado à COVID-19, incluindo o medo da contaminação e da morte, podendo ser utilizada com adaptações em outros momentos de endemias ou pandemias. Não se verificou maior ocorrência de medo entre adolescentes não-heterossexuais, como hipotetizado, contudo, eles revelaram mais engajamento em psicoterapia e mais sintomas de depressão, ansiedade e estresse.

show abstract

“…, v jD ) indicating the latent position of country j in the Euclidean space. When the latent space model is applied to bipartite networks, the size of the Euclidean distance describes the intensity of connection between the two types of nodes, sometimes outside of the social friendship context (see Friel et al, 2016;Wang, 2021;O'Neale et al, 2019;Sarkar and Moore, 2005). When two nodes are closely connected, the Euclidean distance between their latent positions is small.…”

Section: Latent Space Modelmentioning

confidence: 99%

Resilience to stress in bipartite networks: application to the Islamic State recruitment network

Wang

Edgerton

2022

Journal of Complex Networks

Self Cite

View full text Add to dashboard Cite

Networks are resilient to internal failures or external attacks. The resiliency is often beneficial, but there are scenarios where the collapse of a network would benefit society, such as the dismantlement of terrorist, rebel or organized crime groups. In this article, we develop a method to estimate the effect of knockouts and apply our method to the Islamic State recruitment network. Using our novel application, we demonstrate how coordinated attacks against recruiters might reduce the Islamic State’s ability to mobilize new fighters. This analysis has direct implications for research on network resilience and terrorist recruitment.

show abstract

Recent Integrations of Latent Variable Network Modeling With Psychometric Models

Cited by 5 publications

References 76 publications

Latent Variable Machine Learning Framework for Catalysis: General Models, Transfer Learning, and Interpretability

Latent Variable Machine Learning Framework for Catalysis: General Models, Transfer Learning, and Interpretability

Propriedades Psicométricas da Fear of COVID-19 Scale e Avaliação de Medo em Adolescentes Brasileiros

Resilience to stress in bipartite networks: application to the Islamic State recruitment network

Contact Info

Product

Resources

About