Tightly coupled multi-compartment epidemic models tend to expose a weakness of standard Physics-Informed Neural Networks (PINNs). When collocation points are placed uniformly, the network spends much of its capacity on smooth regions while leaving the sharp transients poorly resolved. We work around this by pairing Failure-Informed PINNs (FI-PINNs) with a Self-Adaptive Importance Sampling (SAIS) refinement strategy, and we apply the combination to a nine-equation Susceptible/Vaccinated/Exposed/Infected/Recovered (SVEIR) model that follows three viral strains together with a vaccination compartment. The idea behind the construction is simple. We build a residual-based limit-state function, estimate the failure probability associated with it, and let the network steer its own sampling toward the time intervals where the governing equations are not yet satisfied to within a prescribed tolerance. On the same temporal domain, with identical initial conditions and the same network architecture, SAIS-enhanced FI-PINNs reach a relative L_2 error of 1.04× 10^-6 , against 7.36× 10^-3 for uniform sampling and 3.42× 10^-4 for Residual-Based Adaptive Refinement (RAR) [Lu et. al,. 63:208-228, 2021], an improvement of two to three orders of magnitude. The benefits go beyond raw accuracy. The failure-probability estimate gives an interpretable stopping criterion, and the truncated Gaussian proposal of SAIS keeps the sampling-bias risk under control, a risk that affects purely greedy residual-based schemes. The framework is general enough to be transported to other coupled compartmental systems and to data-assimilation settings in which partial observations are available. 92D30 , 65L05 , 68T07 , 00A71

Prostate cancer is a hormone-dependent cancer characterized by two types of cancer cells, androgen-dependent cancer cells and androgen-resistant ones. The objective of this paper is to present a novel mathematical model for the treatment of prostate cancer under combined hormone therapy and brachytherapy. Using a system of partial differential equations, we quantify and study the evolution of the different cell densities involved in prostate cancer and their responses to the two treatments. Numerical simulations of tumor growth under different therapeutic strategies are explored and presented. The numerical simulations are carried out on FreeFem++ using a 2D finite element method.

Objective: This study examines the integration of climate transition risks into the management of Moroccan equity portfolios, aligning investment strategies with the Sustainable Development Goals (SDGs). It aims to explore approaches that balance sustainability objectives and financial performance while addressing the unique challenges of emerging markets. Theoretical Framework: The research builds on portfolio optimization theories and sustainable finance principles, incorporating climate risk management within the context of the SDGs. A multi-objective evolutionary optimization framework is employed to analyze the trade-offs between carbon intensity (CI) reduction, Tracking Error (TE), and market performance. Method: The study evaluates two strategies: a passive approach aimed at minimizing TE and CI, and an active approach focusing on outperforming the market while prioritizing companies prepared for the energy transition. The analysis highlights data limitations and proposes actionable steps to integrate climate risks into decision-making. Results and Discussion: Findings indicate that a CI reduction exceeding 30% can be achieved with a TE below 0.2%, allowing investors to meet sustainability targets without significant market deviation. The active strategy demonstrates potential for long-term resilience and alignment with SDG objectives, particularly in fostering low-carbon investment opportunities. Research Implications: This study provides practical insights for institutional investors in emerging markets, emphasizing the importance of flexible, data-driven techniques to address climate transition risks. It underscores the need for robust tools and data improvements to support sustainable finance and the achievement of SDGs in Morocco. Originality/Value: This research is among the first to focus on quantifying and integrating climate transition risks into Moroccan equity portfolios. Its innovative use of multi-objective optimization and emphasis on SDGs provide a novel framework for sustainable investment strategies in emerging markets.

We surveyed to measure the satisfaction of policyholders in Morocco, and the results clearly show that the majority of customers do not appreciate the current services. They suffer from the ambiguity of contracts, and delays in reimbursement and do not feel the real impact of insurance in society. To solve this problem, we propose an innovative insurance based on blockchain and waqf. We suggest in this paper, to use smart contracts to create an efficient and automatic process in the collection of premiums and reimbursement of policyholders. The goal of this paper is to build insurance that reflects the true meaning of solidarity through Waqf while integrating transparency and speed through Fintech. This insurance model is supposed to be resilient in times of crisis, have a strong social impact, and be attractive to customers. Many advantages of the proposed model are discussed in the paper. In addition, the suggested insurance model will be represented through simulations on the NetLogo platform. We carry out the analysis in normal times and evaluate the behavior of policyholders in choosing a specific type of insurance, depending on some decision-making tools. We also analyze the impact of insurance during a time of crisis, as a particular example, the crisis experienced during the coronavirus pandemic. The simulations aim to evaluate the model in different situations and prove its efficiency.

This paper provides a comprehensive exploration of physics-informed neural networks and their core features. It delves into their role in tackling inverse problems inherent in ordinary differential equation-based models. Within this context, we introduce a two-group epidemiological model, elucidating its fundamental attributes. The central objective of this research is to accurately estimate the model parameters for both groups in the epidemiological model. We offer a detailed exposition of the adopted methodology, providing insights into the algorithm and the techniques employed for its implementation. Through this analysis, we illuminate the complexities of our study, contributing to the growing body of knowledge in this field, which intersects epidemiology and neural network-based parameter estimation for an enriched understanding of infectious disease dynamics.

In this work, we suggest studying the barriers that prevent from using blockchain technology and smart contracts in the insurance sector. It is possible to improve many services, by introducing ""Fintech"" information technologies which will ensure maximum transparency and speed. The goal of our paper is to answer two main questions: What obstacles stand in the way of the successful use of blockchain technology throughout the insurance sector? Which of them are the most notable obstacles that require decision-makers consideration?. We opt for an analysis of the barriers to blockchain adoption using fuzzy logic for the following reasons. In many realistic situations, it is difficult to gather the exact assessment data; the assessment is based mainly on the decision makers’ knowledge and their experiences using linguistic terms or sentences in a natural or artificial language. The idea is to transform the linguistic variables into fuzzy sets using appropriate membership functions. In other words, fuzzy logic allows a better representation of the uncertainty and subjectivity of decision-makers. In our study, we analyze the answers of twenty experts, - highly skilled professionals with advanced knowledge acquired through education and experience-, about the most significant barriers to blockchain adoption in an interval-valued intuitionistic fuzzy environment. Then, by making use of decision-making tools such as IVIF TOPSIS, we make a ranking of barriers according to their importance to find the most important factors that influence the adoption of blockchain technology. This study’s goal is to propose a model for identifying and tracking the crucial elements that influence managers’ decisions on whether to adopt a financial technology like blockchain in their businesses or not. In the end, we conclude with some recommendations and suggestions to overcome the most important barriers and face future challenges.

Cancer stands as the foremost global cause of mortality, with millions of new cases diagnosed each year. Many research papers have discussed the potential benefits of Machine Learning (ML) in cancer prediction, including improved early detection and personalized treatment options. The literature also highlights the challenges facing the field, such as the need for large and diverse datasets as well as interpretable models with high performance. The aim of this paper is to suggest a new approach in order to select and assess the generalization performance of ML models in cancer prediction, particularly for datasets with limited size. The estimates of the generalization performance are generally influenced by numerous factors throughout the process of training and testing. These factors include the impact of the training–testing ratio as well as the random selection of datasets for training and testing purposes.

Objective: This study evaluates climate transition risks in Morocco, focusing on the energy sector under the climate actions outlined by the Sustainable Development Goals (SDGs). It aims to understand the financial implications of climate transition risk in the Moroccan stock market. Theoretical Framework: The study builds on the frameworks of transition risk assessment and integrated assessment models, employing a Poisson jump process to capture uncertainties in climate policy transitions. Method: A model-based approach is utilized, incorporating CO2 data, integrated assessment models, and six climate scenarios. The analysis uses Monte Carlo simulations to explore potential policy shifts and their impact on the Moroccan stock market. Results and Discussion: Findings reveal significant differences in the resilience of energy sub-sectors, with fossil fuel sectors facing substantial financial losses, while renewable sectors show greater resilience. A responsible investment strategy, prioritizing low-carbon sectors, demonstrates a significant reduction in financial risks compared to a capitalistic alternative. The study highlights the critical role of sustainable investment in mitigating climate transition risks. Research Implications: The results provide valuable insights for Moroccan financial entities, emphasizing the importance of aligning investments with SDGs to minimize financial risks. The findings also offer a foundation for developing climate risk assessment tools that support sustainable finance in Morocco. Originality/Value: This study is the first to focus on quantifying climate transition risk in the Moroccan stock market. It stands out for its innovative approach to modeling the uncertainty associated with climate policies, providing a more refined understanding of how policy shifts impact financial risks.

This work has two principal goals. First, we investigate the asymptotic behavior of a two-group epidemiological model and determine the expression of its basic reproduction number using the dynamical systems approach based on the spectral radius of the relative matrix. Second, we simulate the obtained analytical results using a new deep learning method that associates the ordinary differential equations governing the model to neural networks. A general disease-free equilibrium is considered and sufficient conditions of stability and convergence are formulated. A detailed description of the neural network model used in the simulation is provided. Moreover, the proposed deep learning simulation algorithm is compared to the simulation provided by "odeint", a function from "SciPy" which is a Python library of mathematical routines.

This paper bring insight on two predominant paradigms in economic modelling. It aims, first, at providing a general view on the literature contrasting Dynamic Stochastic General Equilibrium (DSGE) and Agent Based Models (ABM). And second, at showing the differences between these two types of modeling using two models developed in previous papers [1–3]. The paper discusses questions related to the improvement of modern theory, the understanding of economic and financial phenomena and the limits of the different models. It presents a short literature review and reveal a worthy part of the debate made on theoretical framework in macroeconomics and the different way of thought. The paper analyzes the current dominant models namely DSGE and examines if a different methodological approach could be beneficial.

The aim of this work is to study the dynamics of viral infection by a mathematical model using a differential equation with a single delay corresponding to the duration of proliferation and differentiation of immune cells and the time required to program activated CTLs. Asymptotic and global stability conditions for the considered delayed differential equation are defined in order to study the asymptotic behaviour of the solutions. Key theorems are proven using the theory of monotone dynamical systems, mainly the results established by M. Pituk in 2003. Sufficient conditions of stability of the nonzero equilibrium have been established and formulated in terms of the efficiency and delay of the immune response. Numerical simulations of the model are given to validate analytical results.

The purpose of this work is the study of the qualitative behavior of the homogeneous in space solution of a delay differential equation arising from a model of infection dynamics. This study is mainly based on the monotone dynamical systems theory. Existence and smoothness of solutions are proved, and conditions of asymptotic stability of equilibriums in the sense of monotone dynamical systems are formulated. Then, sufficient conditions of global stability of the nonzero steady state are derived, for the two typical forms of the function f, specifying the efficiency of immune response-mediated virus elimination. Numerical simulations illustrate the analytical results. The obtained theoretical results have been applied, in a context of COVID-19 data calibration, to forecast the immunological behaviour of a real patient.

In this paper, two different approaches for tumor growth modeling are presented and implemented. In the first part of the paper, a macroscopic approach using a PDE model, where the tumor is viewed as a cell mass, is implemented using the level-set method to track the tumor moving boundary in one hand by using Darcy’s law to compute the normal velocity of the free boundary and on the other hand using the shape optimization to draw the normal velocity. In the second part of the paper, a microscopic approach, which focuses on the cellular scale, is presented. A hybrid model using agent-based modeling for the cell behavior and a PDE for the description of the tumor environment is presented. A sensitivity analysis is performed on the hybrid model for a better understanding of its impact on the tumor growth. Numerical experiments are provided for the proposed approaches.

This paper suggests an enhanced machine-learning-based system to guide future stock price decisions. In reality, most existing machine learning systems, such as SEA (Stream Ensemble Algorithm), VFDT (Very Fast Decision Tree ), and online bagging and boosting, keep models updated with only new data and reduce training timeframes to allow working rapidly with the most recent model. However, limited learning times and the exclusion of essential information from previous data may result in a bad performance. When it comes to learning models, our system takes a different approach. It builds several models based on random selections of historical data from the main stock as well as related stocks. The best models are then combined to generate a final, performant model. We performed an empirical study on five Islamic stock market indices. We can say from the results that our system outperforms the existing published algorithms. This framework can contribute then to having an enhanced system that will enable different stakeholders to make rapid decisions based on the forecasted trend of indices.