Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Nikolai Sergeevich Akintsov has a total of 19 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Entropy-Based Detection of the Rest-to-Firing Transition in Morris–Lecar Dynamics Reconstructed from Sparse Voltage Data by Physics-Informed Neural Networks

    by Nikolai S. Akintsov, Vladislav S. Igumnov

    Entropy measures are widely used to detect changes of dynamical regime, but their reliability when the signal must first be reconstructed from sparse, noisy recordings is rarely tested against an independently known transition. We studied the rest-to-firing transition of the class-I Morris–Lecar neuron, located by continuation at a saddle-node on invariant circle bifurcation (39.963 μA cm−2), and scored sample, permutation, weighted permutation, and spectral entropy by their localization error. Voltage traces were reduced to 50–5% of their samples, corrupted by noise, and reconstructed by a physics-informed neural network (PINN), cubic splines, and an ODE least-squares fit. On reference signals, permutation-type and spectral entropy located the transition to the grid resolution (0.012 μA cm−2), whereas sample entropy erred by 0.21 μA cm−2. Without noise, all reconstructions preserved these results down to 5% sparsity, and the PINN recovered the unobserved recovery variable with a median relative error of 0.027. With noise, residual fluctuations in resting windows reversed the entropy contrast at the transition while the localization error often stayed small; the effect was strongest for splines and absent for the ODE fit. Localization error must therefore be reported together with regime separability, and the quiescent threshold is part of the measurement protocol.

  • Relativistic electron dynamics in Gaussian laser fields: coupled-parameter theory, multi-regime analysis, and physics-informed neural network modeling

    by Nikolai S Akintsov, Artem P Nevecheria, Stepan N Andreev, Qinghua Qin

    We present a treatment of relativistic electron dynamics in a spatially and temporally Gaussian laser pulse in 3+1 dimensions, applied to the design of a compact laser synchrotron source (LSS). The framework is the method of coupled parameters within Lobachevsky (hyperbolic) geometry, in which all kinematic quantities—momentum, energy, velocity, and coordinates—are expressed as simultaneous hyperbolic functions of the rapidity θ , the unique parameter for which the energy–momentum Lorentz invariant E 2 − P 2 = 1 holds identically. Lagrangian and Hamiltonian formulations are derived, with closed-form asymptotics for the nonrelativistic ( θ ≪ 1 ) and ultrarelativistic ( θ ≫ 1 ) limits. The interaction is analyzed across three intensity regimes ( I = 10 18 , 10 20 , 10 22 W cm − 2 : linear Thomson, weakly nonlinear, and strongly relativistic); trajectories, phase portraits, and spectral power distributions are computed analytically for each. Phase-synchronization conditions follow from the equations of motion; the stabilizing role of a constant magnetic field is recovered in closed form as the classical cyclotron-autoresonance mechanism, which breaks the Lawson–Woodward restriction. We quantitatively delimit the validity of the radiation-reaction-free classical description in the ( a 0 , γ 0 ) plane: for an electron starting from rest the quantum-nonlinearity parameter is χ ≲ 2 × 10 − 4 and the radiated-energy fraction is R ≲ 10 − 2 even at a 0 = 85 , so radiation reaction is negligible, whereas the counter-propagating beam geometry relevant to a LSS reaches χ ∼ 1 and lies outside the classical model. A physics-informed neural network, with the Lorentz invariant embedded as a soft training constraint, enables rapid exploration of the ( a 0 , H 0 ) design space, reducing the typical invariant violation by one to four orders of magnitude relative to an unconstrained baseline and delivering a ∼ 430 × inference-time speedup over an eighth-order Runge–Kutta (DOP853) reference. The contributions are thus methodological and computational: a closed-form unification of the three intensity regimes and an open, reproducible, artificial-intelligence-assisted toolkit for compact laser-synchrotron-source design.

  • Open Access

    Symmetry-Preserving Physics-Informed Neural Network Framework for Relativistic Charged-Particle Dynamics in 3+1 Dimensions

    by Nikolai S. Akintsov, Artem P. Nevecheria, Gaoteng Yuan, Vladislav S. Igumnov, Stepan N. Andreev, Qing-Hua Qin

    Standard pushers for the relativistic equations of motion of a charged particle in an electromagnetic field—Boris, Vay, Higuera–Cary—do not, in general, preserve the full symplectic structure of the underlying Hamiltonian system, while high-order non-symplectic schemes such as Runge–Kutta accumulate secular error over long times. We propose a two-stage, symmetry-preserving framework (SP-PINN) for the 3+1-dimensional relativistic dynamics of a charged particle in a prescribed field, including a focused Gaussian laser pulse, that pairs a physics-informed neural network with an explicit symplectic integrator: the network learns a surrogate relativistic Hamiltonian, while the integrator—which is not itself learned—advances it. In Stage 1, an unsupervised physics-informed neural network learns the surrogate from the covariant equations of motion using a Lorentz-invariant loss that enforces the mass-shell constraint H=mc2γ; in Stage 2, the surrogate is advanced with an explicit symplectic map built on Tao’s extended phase space, valid for the non-separable relativistic Hamiltonian. To isolate the geometric integrator from neural-network approximation error, every benchmark figure advances the analytic relativistic Hamiltonian through Stage 2, the learned Stage-1 surrogate being assessed separately. We benchmark against the Boris pusher and Runge–Kutta on three core test problems (adding the Higuera–Cary pusher in the symplecticity diagnostic), supplemented by plane-wave, ensemble, and pulse-family studies, and we measure the first Poincaré–Cartan loop invariant directly as a quantitative diagnostic of symplecticity. The magnetic-field test illustrates the contrast between bounded and secular error growth: Runge–Kutta drifts secularly, the Boris pusher conserves the invariants to machine precision as a volume-preserving gyro-integrator, and the symplectic map keeps the error bounded for all time; on a non-integrable magnetic trap, where no exact volume-preserving rotation exists, the symplectic map alone keeps the energy error bounded. The learned surrogate is the current accuracy bottleneck—not yet competitive with the conventional pushers for the static cases—but for the demanding laser case, a vector-potential light-cone reformulation reduces this surrogate error to (3.0±0.1)×10−4 (three seeds) and yields learned trajectories that remain phase-coherent over essentially the whole interaction. The framework targets laser–plasma acceleration, synchrotron-radiation modeling, and particle tracking.


  • Invariant descriptions of classical relativistic particle motion in 3+1 dimensions

    by N. S. Akintsov, A. P. Nevecheria, G. F. Kopytov, S. N. Andreev, Yongjie Yang, Qing-Hua Qin

    A Lorentz-invariant formalism for the classical dynamics of a free relativistic particle in 3+1 dimensions is developed within the framework of Lobachevsky hyperbolic geometry using basis gyrovectors. A modified metric representation of the Lorentz-invariant space-time coordinate is introduced, and the general solution for the projective coordinate function is derived via variation of variables, establishing the connection between the projective velocity components and their canonically conjugate variables through the angular integral of motion. The invariant projections of the particle velocity, momentum, energy, and coordinates onto three basis gyrovectors are obtained explicitly as functions of the rapidity. A systematic analysis reveals three distinct kinematic regimes for the transverse momentum: it is absent at low rapidities, real and diverging at an upper boundary in the intermediate regime, and transitions into the imaginary domain in the ultrarelativistic limit, indicating a change of geometry of momentum space. The projective velocity attains a maximum at a finite rapidity value — a physical analog of the gyrosynchrotron radiation regime. The geometric structure of the particle trajectory in the coordinate plane is analyzed and its conformal mapping onto the unit circle is demonstrated. As a benchmark application, the formalism is applied to a relativistic charged particle in the field of a linearly polarized plane electromagnetic wave, reducing the derivation of transverse momentum to a single projection step versus four sequential steps required by the standard Hamilton–Jacobi approach. Results agree exactly with the known classical solution.


  • Open Access

    Symmetry-Preserving Physics-Informed Neural Network Framework for Relativistic Charged-Particle Dynamics in 3+1 Dimensions

    by Nikolai S. Akintsov, Artem P. Nevecheria, Gaoteng Yuan, Vladislav S. Igumnov, Stepan N. Andreev, Qing-Hua Qin

    Standard pushers for the relativistic equations of motion of a charged particle in an electromagnetic field—Boris, Vay, Higuera–Cary—do not, in general, preserve the full symplectic structure of the underlying Hamiltonian system, while high-order non-symplectic schemes such as Runge–Kutta accumulate secular error over long times. We propose a symmetry-preserving physics-informed neural network framework (SP-PINN) for the 3+1-dimensional relativistic dynamics of a charged particle in a prescribed field, including a focused Gaussian laser pulse. The method is two-stage: an unsupervised physics-informed neural network learns a surrogate relativistic Hamiltonian from the covariant equations of motion using a Lorentz-invariant loss that enforces the mass-shell constraint H=mc2γ; the surrogate is then advanced with an explicit symplectic map built on Tao’s extended phase space, valid for the non-separable relativistic Hamiltonian. We benchmark against the Boris pusher and Runge–Kutta on three test problems. The magnetic-field test illustrates the contrast between bounded and secular error growth: Runge–Kutta drifts secularly, the Boris pusher conserves the invariants to machine precision as a volume-preserving gyro-integrator, and the symplectic map keeps the error bounded for all time; on a non-integrable magnetic trap, where no exact volume-preserving rotation exists, the symplectic map alone keeps the energy error bounded. The learned surrogate is the current accuracy bottleneck; for the demanding laser case a vector-potential light-cone reformulation reduces its error to (3.0±0.1)×10−4 (three seeds) and yields learned trajectories that remain phase-coherent over essentially the whole interaction. The framework targets laser–plasma acceleration, synchrotron-radiation modeling, and particle tracking.

  • Open Access

    Dynamics of relativistic particles in an ion–cyclotron trap in the presence of an external ultrashort laser pulse

    by Nikolai S. Akintsov, Artem P. Nevecheria, Stepan N. Andreev, Gennady F. Kopitov

    The study explores the dynamics of relativistic particles in an ion–cyclotron trap under ultrashort laser pulses with intensities around 1022 W/cm2. The focus is on the trajectory, speed, momentum, and energy of the particles in the ion trap, aiming to determine the optimal electric and magnetic field intensities for effective confinement. The effects of Gaussian laser pulses and the ion trap field are analyzed in the context of achieving cyclotron autoresonance. Numerical solutions of relativistic particle motion, including methods such as Boris, GIGI2, and Runge–Kutta, are compared. The results demonstrate the complementary nature of analytical solutions in Lobachevskian geometry and numerical solutions in Euclidean geometry.


  • Open Access

    Special Relativity in Terms of Hyperbolic Functions with Coupled Parameters in 3+1 Dimensions

    by Nikolai S. Akintsov, Artem P. Nevecheria, Gennadii F. Kopytov, Yongjie Yang, Tun Cao

    This paper presents a method for parameterizing new Lorentz spacetime coordinates based on coupled parameters. The role of symmetry in rapidity in special relativity is explored, and invariance is obtained for new spacetime intervals with respect to the Lorentz transformation. Using the Euler–Hamilton equations, an additional angular rapidity and perpendicular rapidity are obtained, and the Hamiltonian and Lagrangian of a relativistic particle are expanded into rapidity spectra. A so-called passage to the limit is introduced that makes it possible to decompose physical quantities into spectra in terms of elementary functions when explicit decomposition is difficult. New rapidity-dependent Lorentz spacetime coordinates are obtained. The descriptions of particle motion using the old and new Lorentz spacetime coordinates as applied to plane laser pulses are compared in terms of the particle kinetic energy. Based on a classical model of particle motion in the field of a plane monochromatic electromagnetic wave and that of a plane laser pulse, rapidity-dependent spectral decompositions into elementary functions are presented, and the Euler–Hamilton equations are derived as rapidity functions in 3+1 dimensions. The new and old Lorentz spacetime coordinates are compared with the Fermi spacetime coordinates. The proper Lorentz groups SO(1,3) with coupled parameters using the old and new Lorentz spacetime coordinates are also compared. As a special case, the application of Lorentz spacetime coordinates to a relativistic hydrodynamic system with coupled parameters in 1+1 dimensions is demonstrated.


  • Open Access

    Special Relativity in Terms of Hyperbolic Functions with Coupled Parameters in 3+1 Dimensions

    by Nikolai S. Akintsov, Artem P. Nevecheria, Gennadii F. Kopytov, Yongjie Yang

    This paper presents a new method for parameterizing the Lorentz group based on coupled parameters. From the Euler–Hamilton equations, an additional angular rapidity and perpendicular rapidity are obtained, and the Hamiltonian and Lagrangian of a relativistic particle are expanded into rapidity spectra. A so-called passage to the limit is introduced that makes it possible to decompose physical quantities into spectra in terms of elementary functions when explicit decomposition is difficult. New rapidity-dependent Lorentz-invariant coordinates are obtained, and the descriptions of particle motion using the old and new Lorentz-invariant forms as applied to plane waves are compared. Based on a classical model of particle motion in the field of a plane monochromatic electromagnetic wave and in that of a plane laser pulse, the rapidity-dependent spectral decompositions into elementary functions are presented, and the Euler–Hamilton equations are derived as rapidity functions in 3+1 dimensions.


  • Open Access

    Lagrangian and Hamiltonian Formalisms for Relativistic Mechanics with Lorentz-Invariant Evolution Parameters in 1 + 1 Dimensions

    by Nikolai S. Akintsov, Artem P. Nevecheria, Gennadii F. Kopytov, Yongjie Yang

    This article presents alternative Hamiltonian and Lagrangian formalisms for relativistic mechanics using proper time and proper Lagrangian coordinates in 1 + 1 dimensions as parameters of evolution. The Lagrangian and Hamiltonian formalisms for a hypothetical particle with and without charge are considered based on the relativistic equation for the dynamics and integrals of particle motion. A relativistic invariant law for the conservation of energy and momentum in the Lorentz representation is given. To select various generalized coordinates and momenta, it is possible to modify the Lagrange equations of the second kind due to the relativistic laws of conservation of energy and momentum. An action function is obtained with an explicit dependence on the velocity of the relativistic particles. The angular integral of the particle motion is derived from Hamiltonian mechanics, and the displacement Hamiltonian is obtained from the Hamilton–Jacobi equation. The angular integral of the particle motion θ is an invariant form of the conservation law. It appears only at relativistic intensities and is constant only in a specific case. The Hamilton–Jacobi–Lagrange equation is derived from the Hamilton–Jacobi equation and the Lagrange equation of the second kind. Using relativistic Hamiltonian mechanics, the Euler–Hamilton equation is obtained by expressing the energy balance through the angular integral of the particle motion θ. The given conservation laws show that the angular integral of the particle motion reflects the relativistic Doppler effect for particles in 1 + 1 dimensions. The connection between the integrals of the particle motion and the doubly special theory of relativity is shown. As an example of the applicability of the proposed invariant method, analyses of the motion of relativistic particles in circularly polarized, monochromatic, spatially modulated electromagnetic plane waves and plane laser pulses are given, and comparisons are made with calculations based on the Landau and Lifshitz method. To allow for the analysis of the oscillation of a particle in various fields, a phase-plane method is presented.