-
Quantum Recurrent Neural Networks with Encoder-Decoder for Time-Dependent Partial Differential Equations
Authors:
Yuan Chen,
Abdul Khaliq,
Khaled M. Furati
Abstract:
Nonlinear time-dependent partial differential equations are essential in modeling complex phenomena across diverse fields, yet they pose significant challenges due to their computational complexity, especially in higher dimensions. This study explores Quantum Recurrent Neural Networks within an encoder-decoder framework, integrating Variational Quantum Circuits into Gated Recurrent Units and Long…
▽ More
Nonlinear time-dependent partial differential equations are essential in modeling complex phenomena across diverse fields, yet they pose significant challenges due to their computational complexity, especially in higher dimensions. This study explores Quantum Recurrent Neural Networks within an encoder-decoder framework, integrating Variational Quantum Circuits into Gated Recurrent Units and Long Short-Term Memory networks. Using this architecture, the model efficiently compresses high-dimensional spatiotemporal data into a compact latent space, facilitating more efficient temporal evolution. We evaluate the algorithms on the Hamilton-Jacobi-Bellman equation, Burgers' equation, the Gray-Scott reaction-diffusion system, and the three dimensional Michaelis-Menten reaction-diffusion equation. The results demonstrate the superior performance of the quantum-based algorithms in capturing nonlinear dynamics, handling high-dimensional spaces, and providing stable solutions, highlighting their potential as an innovative tool in solving challenging and complex systems.
△ Less
Submitted 18 February, 2025;
originally announced February 2025.
-
Rational Approximations for Oscillatory Two-Parameter Mittag-Leffler Function
Authors:
Aljowhara H. Honain,
Khaled M. Furati,
Ibrahim O. Sarumi,
Abdul Q. M. Khaliq
Abstract:
The two-parameter Mittag-Leffler function $E_{α, β}$ is of fundamental importance in fractional calculus. It appears frequently in the solutions of fractional differential and integral equations. Nonetheless, this vital function is often expensive to compute. Several attempts have been made to construct cost-effective and accurate approximations. These attempts focus mainly on the completely monot…
▽ More
The two-parameter Mittag-Leffler function $E_{α, β}$ is of fundamental importance in fractional calculus. It appears frequently in the solutions of fractional differential and integral equations. Nonetheless, this vital function is often expensive to compute. Several attempts have been made to construct cost-effective and accurate approximations. These attempts focus mainly on the completely monotone Mittag-Leffler functions. However, when $α> 1$ the monotonicity property is largely lost and as such roots and oscillations are exhibited. Consequently, existing approximants constructed mainly for $α\in (0,1)$ often fail to capture this oscillatory behavior. In this paper, we construct computationally efficient and accurate rational approximants for $E_{α, β}(-t)$, $t \ge 0$, with $α\in (1,2)$. This construction is fundamentally based on the decomposition of Mittag-Leffler function with real roots into one without and a polynomial. Following which new approximants are constructed by combining the global Padé approximation with a polynomial of appropriate degree. The rational approximants are extended to approximation of matrix Mittag-Leffler and different approaches to achieve efficient implementation for matrix arguments are discussed. Numerical experiments are provided to illustrate the significant accuracy improvement achieved by the proposed approximants.
△ Less
Submitted 12 December, 2023;
originally announced December 2023.
-
Fourier Spectral Methods with Exponential Time Differencing for Space-Fractional Partial Differential Equations in Population Dynamics
Authors:
A. P. Harris,
T. A. Biala,
A. Q. M. Khaliq
Abstract:
Physical laws governing population dynamics are generally expressed as differential equations. Research in recent decades has incorporated fractional-order (non-integer) derivatives into differential models of natural phenomena, such as reaction-diffusion systems. In this paper, we develop a method to numerically solve a multi-component and multi-dimensional space-fractional system. For space disc…
▽ More
Physical laws governing population dynamics are generally expressed as differential equations. Research in recent decades has incorporated fractional-order (non-integer) derivatives into differential models of natural phenomena, such as reaction-diffusion systems. In this paper, we develop a method to numerically solve a multi-component and multi-dimensional space-fractional system. For space discretization, we apply a Fourier spectral method that is suited for multidimensional PDE systems. Efficient approximation of time-stepping is accomplished with a locally one dimensional exponential time differencing approach. We show the effect of different fractional parameters on growth models and consider the convergence, stability, and uniqueness of solutions, as well as the biological interpretation of parameters and boundary conditions.
△ Less
Submitted 6 December, 2022;
originally announced December 2022.
-
Identification and prediction of time-varying parameters of COVID-19 model: a data-driven deep learning approach
Authors:
Jie Long,
Abdul Khaliq,
Khaled Furati
Abstract:
Data-driven deep learning provides efficient algorithms for parameter identification of epidemiology models. Unlike the constant parameters, the complexity of identifying time-varying parameters is largely increased. In this paper, a variant of physics-informed neural network (PINN) is adopted to identify the time-varying parameters of the Susceptible-Infectious-Recovered-Deceased model for the sp…
▽ More
Data-driven deep learning provides efficient algorithms for parameter identification of epidemiology models. Unlike the constant parameters, the complexity of identifying time-varying parameters is largely increased. In this paper, a variant of physics-informed neural network (PINN) is adopted to identify the time-varying parameters of the Susceptible-Infectious-Recovered-Deceased model for the spread of COVID-19 by fitting daily reported cases. The learned parameters are verified by utilizing an ordinary differential equation solver to compute the corresponding solutions of this compartmental model. The effective reproduction number based on these parameters is calculated. Long Short-Term Memory (LSTM) neural network is employed to predict the future weekly time-varying parameters. The numerical simulations demonstrate that PINN combined with LSTM yields accurate and effective results.
△ Less
Submitted 17 March, 2021;
originally announced March 2021.
-
One-sided Shewhart control charts for monitoring the ratio of two normal variables in Short Production Runs
Authors:
K. D. Tran,
Q. U. A Khaliq,
A. A. Nadi,
H Tran,
K. P. Tran
Abstract:
Monitoring the ratio of two normal random variables plays an important role in several manufacturing environments. For short production runs, however, the control charts assumed infinite processes cannot function effectively to detect anomalies. In this paper, we tackle this problem by proposing two one-sided Shewhart-type charts to monitor the ratio of two normal random variables for an infinite…
▽ More
Monitoring the ratio of two normal random variables plays an important role in several manufacturing environments. For short production runs, however, the control charts assumed infinite processes cannot function effectively to detect anomalies. In this paper, we tackle this problem by proposing two one-sided Shewhart-type charts to monitor the ratio of two normal random variables for an infinite horizon production. The statistical performance of the proposed charts is investigated using the truncated average run length as a performance measure in short production runs. In order to help the quality practitioner to implement these control charts, we have provided ready-to-use tables of the control limit parameters. An illustrative example from the food industry is given for illustration.
△ Less
Submitted 3 October, 2020;
originally announced October 2020.
-
How Efficient is Contact Tracing in Mitigating the Spread of Covid-19? A Mathematical Modeling Approach
Authors:
T. A. Biala,
Y. O. Afolabi,
A. Q. M. Khaliq
Abstract:
Contact Tracing (CT) is one of the measures taken by government and health officials to mitigate the spread of the novel coronavirus. In this paper, we investigate its efficacy by developing a compartmental model for assessing its impact on mitigating the spread of the virus. We describe the impact on the reproduction number ${\mathcal{R}_0}$ of COVID-19. In particular, we discuss the importance a…
▽ More
Contact Tracing (CT) is one of the measures taken by government and health officials to mitigate the spread of the novel coronavirus. In this paper, we investigate its efficacy by developing a compartmental model for assessing its impact on mitigating the spread of the virus. We describe the impact on the reproduction number ${\mathcal{R}_0}$ of COVID-19. In particular, we discuss the importance and relevance of parameters of the model such as the number of reported cases, effectiveness of tracking and monitoring policy, and the transmission rates to contact tracing. We describe the terms "perfect tracking", "perfect monitoring" and "perfect reporting" to indicate that traced contacts will be tracked while incubating, tracked contacts are efficiently monitored so that they do not cause secondary infections, and all infected persons are reported, respectively. We consider three special scenarios: (1) perfect monitoring and perfect tracking of contacts of a reported case, (2) perfect reporting of cases and perfect monitoring of tracked reported cases and (3) perfect reporting and perfect tracking of contacts of reported cases. Furthermore, we gave a lower bound on the proportion of contacts to be traced to ensure that the effective reproduction, ${\mathcal{R}_c}$, is below one and describe ${\mathcal{R}_c}$ in terms of observable quantities such as the proportion of reported and traced cases. Model simulations using the COVID-19 data obtained from John Hopkins University for some selected states in the US suggest that even late intervention of CT may reasonably reduce the transmission of COVID-19 and reduce peak hospitalizations and deaths.
△ Less
Submitted 20 November, 2021; v1 submitted 9 August, 2020;
originally announced August 2020.
-
Highly Accurate Global Padé Approximations of Generalized Mittag-Leffler Function and its Inverse
Authors:
Ibrahim O. Sarumi,
Khaled M. Furati,
Abdul Q. M. Khaliq
Abstract:
The two-parametric Mittag-Leffler function (MLF), $E_{α,β}$, is fundamental to the study and simulation of fractional differential and integral equations. However, these functions are computationally expensive and their numerical implementations are challenging. In this paper, we present a unified framework for developing global rational approximants of $E_{α,β}(-x)$, $x>0$, with…
▽ More
The two-parametric Mittag-Leffler function (MLF), $E_{α,β}$, is fundamental to the study and simulation of fractional differential and integral equations. However, these functions are computationally expensive and their numerical implementations are challenging. In this paper, we present a unified framework for developing global rational approximants of $E_{α,β}(-x)$, $x>0$, with $\{ (α,β): 0 < α\leq 1, β\geq α, (α,β) \ne (1,1) \}$. This framework is based on the series definition and the asymptotic expansion at infinity. In particular, we develop three types of fourth-order global rational approximations and discuss how they could be used to approximate the inverse function. Unlike existing approximations which are either limited to MLF of one parameter or of low accuracy for the two-parametric MLF, our rational approximants are of fourth order accuracy and have low percentage error globally. For efficient utilization, we study the partial fraction decomposition and use them to approximate the two-parametric MLF with a matrix argument which arise in the solutions of fractional evolution differential and integral equations.
△ Less
Submitted 23 December, 2019;
originally announced December 2019.
-
Slow-scale split-step tau-leap method for stiff stochastic chemical systems
Authors:
Viktor Reshniak,
Abdul Khaliq,
David Voss
Abstract:
Tau-leaping is a family of algorithms for the approximate simulation of the discrete state continuous time Markov chains. Motivation for the development of such methods can be found, for instance, in the fields of chemical kinetics and systems biology. It is known that the dynamical behavior of biochemical systems is often intrinsically stiff representing a serious challenge for their numerical ap…
▽ More
Tau-leaping is a family of algorithms for the approximate simulation of the discrete state continuous time Markov chains. Motivation for the development of such methods can be found, for instance, in the fields of chemical kinetics and systems biology. It is known that the dynamical behavior of biochemical systems is often intrinsically stiff representing a serious challenge for their numerical approximation. The naive extension of stiff deterministic solvers to stochastic integration often yields numerical solutions with either impractically large relaxation times or incorrectly resolved covariance. In this paper, we propose a splitting heuristic which helps to resolve some of these issues. The proposed integrator contains a number of unknown parameters which are estimated for each particular problem from the moment equations of the corresponding linearized system. We show that this method is able to reproduce the exact mean and variance of the linear scalar test equation and demonstrates a good accuracy for the arbitrarily stiff systems at least in the linear case. The numerical examples for both linear and nonlinear systems are also provided and the obtained results confirm the efficiency of the considered splitting approach.
△ Less
Submitted 18 December, 2018; v1 submitted 23 September, 2017;
originally announced September 2017.
-
Split-step Milstein methods for multi-channel stiff stochastic differential systems
Authors:
V. Reshniak,
A. Q. M. Khaliq,
D. A. Voss,
G. Zhang
Abstract:
We consider split-step Milstein methods for the solution of stiff stochastic differential equations with an emphasis on systems driven by multi-channel noise. We show their strong order of convergence and investigate mean-square stability properties for different noise and drift structures. The stability matrices are established in a form convenient for analyzing their impact arising from differen…
▽ More
We consider split-step Milstein methods for the solution of stiff stochastic differential equations with an emphasis on systems driven by multi-channel noise. We show their strong order of convergence and investigate mean-square stability properties for different noise and drift structures. The stability matrices are established in a form convenient for analyzing their impact arising from different deterministic drift integrators. Numerical examples are provided to illustrate the effectiveness and reliability of these methods.
△ Less
Submitted 25 November, 2014;
originally announced November 2014.
