Skip To Main Content
Dmitriy Tarkhov on Muck Rack

Dmitriy Tarkhov

As seen in: MDPI
Is this you? As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work. Claim your profile
Back to Dmitriy Tarkhov’s profile

Articles by Dmitriy Tarkhov

Adapting PINN Models of Physical Entities to Dynamical Data

|
By Dmitriy Tarkhov, Tatiana Lazovskaya, Valery Antonov
| MDPI
5. Conclusions We have studied the adaptive properties of PINN and PPINN solutions using dynamic real-time data. The ability of the adaptive parametric and classical PINNs to adapt was compared for the first time in this study. This comparison was conducted using problem examples that involved various dependency functions of an unknown dynamic parameter.
Open in Who Shared Issue with article?

Evolutionary PINN Learning Algorithms Inspired by Approximation to Pareto Front for Solving Ill-Posed Problems

|
By Tatiana Lazovskaya, Dmitriy Tarkhov, Maria Chistyakova, Egor Razumov
| MDPI
1. Introduction The vast majority of papers presenting physics-informed neural network models and solutions use a pre-selected network architecture, mentioning that the corresponding parameters were found empirically. This dilemma is also mentioned in Haykin’s classic book [1]. The concept of architecture in this case can include both types of activation functions, the number of layers and neurons in them, and methods for initializing the output parameters of a neural network.
Open in Who Shared Issue with article?

Sensors | Free Full-Text | Constructing Physics-Informed Neural Networks with Architecture Based on Analytical Modification of Numerical Methods by Solving the Problem of Modelling Processes in a Chemical Reactor

|
By Galina Malykhina, Dmitriy Tarkhov, Tatiana Lazovskaya
| MDPI
Abstract : A novel type of neural network with an architecture based on physics is proposed. The network structure builds on a body of analytical modifications of classical numerical methods. A feature of the constructed neural networks is defining parameters of the governing equations as trainable parameters. Constructing the network is carried out in three stages. In the first step, a neural network solution to an equation corresponding to a numerical scheme is constructed.
Open in Who Shared Issue with article?
Show More