arXiv subjects
Publications and source records attributed to V. A. Ilyin.
The purpose of this paper is to review the most popular deep learning methods used to analyze astroparticle data obtained with Imaging Atmospheric Cherenkov Telescopes and provide references to the original papers.
The paper presents a REDUCE program for the simplification of tensor expressions that are considered as formal indexed objects. The proposed algorithm is based on the consideration of tensor expressions as vectors in some linear space. This linear space is formed by all the elements of the group algebra of the corresponding tensor expression. Such approach permits us to simplify the tensor expressions possessing symmetry properties, summation (dummy) indices and multiterm identities by unify manner. The canonical element for the tensor expression is defined in terms of the basic vectors of this linear space. The main restriction of the algorithm is the dimension of the linear space that is equal to N!, where N is a number of indices of the tensor expression. The program uses REDUCE as user interface.
Growth kinetics of AlGaN in NH3 MBE under significant Ga desorption was studied. It was found that the addition of gallium stimulates 2D growth and provides better morphology of films compared to pure AlN. The effect was experimentally observed at up to 98% desorption of the impinging gallium. We found that, under the conditions of significant thermal desorption, larger amounts of gallium were retained at lateral boundaries of 3D surface features than at flat terraces because of the higher binding energy of Ga atoms at specific surface defects. The selective accumulation of gallium resulted in an increase in lateral growth component through the formation of the Ga-enriched AlGaN phase at boundaries of 3D surface features. We studied temperature dependence of AlGaN growth rate and developed a kinetic model analytically describing this dependence. As the model was in good agreement with the experimental data, we used it to estimate the increase in the binding energy of Ga atoms at surface defects compared to terrace surface cites using data on the Ga content in different AlGaN phases. We also applied first-principles calculations to the thermodynamic analysis of stable configurations on the AlN surface and then used these surface configurations to compare the binding energy of Ga atom at terraces and steps. Both first-principles calculations and analytical estimations of the experimental results gave similar values of difference in binding energies; this value is 0.3 eV.
Several systems in the solid state have been suggested as promising candidates for spin-based quantum information processing. In spite of significant progress during the last decade, there is a search for new systems with higher potential [D. DiVincenzo, Nature Mat. 9, 468 (2010)]. We report that silicon vacancy defects in silicon carbide comprise the technological advantages of semiconductor quantum dots and the unique spin properties of the nitrogen-vacancy defects in diamond. Similar to atoms, the silicon vacancy qubits can be controlled under the double radio-optical resonance conditions, allowing for their selective addressing and manipulation. Furthermore, we reveal their long spin memory using pulsed magnetic resonance technique. All these results make silicon vacancy defects in silicon carbide very attractive for quantum applications.
In some supersymmetric extensions of the Standard Model fairly light superpartner of t-quark is predicted, which may form bound states - stoponiums. We estimate potentials of TESLA linear collider in search for stoponium, considering the basic e+e- option and the gamma-gamma option (Photon Linear Collider - PLC). It is found that PLC could be the best machine for discovery of these new narrow strong resonances. It can produce thousands stoponiums per 100 fb$^-1$ integrated luminosity in the high energy peak. In the case of scenarios when stoponium decays mainly into gluons the signal/background ratio is about 1/4. The channel S->hh into two lightest Higgs bosons could be also seen with high significance. Thus, several weeks run is sufficient for the stoponium discovery, if its mass is approximately known (e.g. from observation of direct stops production at LHC). Then, in MSSM scenarios with dominant S->hh decay PLC shows excellent possibilities to discover bound state of stops, practically immediately after beginning of operating. The e+e- option also has some prospects to observe stoponium but only in the case of scenarios with dominant decay into two lightest Higgs bosons, with tens of events per 100 $fb^{-1}$. Interesting possibility appears in the case when the resonance is seated on 0.1% width luminosity peak - one could resolve the stoponium exited states.
We present a method for symbolic evaluation of Feynman amplitudes. We construct special polarization basis for spinor particles which produces compact expressions for tensor products of basis spinors.
CompHEP, as a partonic event generator, and PYTHIA, as a generator of final states of detectable objects, are interfaced. Thus, integrated tool is proposed for simulation of (almost) arbitrary collision processes at the level of detectable particles. Exact (multiparticle) matrix elements, convolution with structure functions, decays, partons hadronization and (optionally) parton shower evolution are basic stages of calculations. The PEVLIB library of event generators for LHC processes is described.
In some supersymmetric extensions of the Standard Model fairly light superpartner of t-quark is predicted, which may form bound states ({\it stoponiums}) under certain conditions. We study prospects of search for stoponium at the future Photon Linear Collider. It is found that this machine could be the best machine for discovery and study of these resonances at some scenarios of supersymmetric extension of the Standard Model. In particular, if the $hh$ decay channel is dominant stoponium could be observed at the beginning of PLC run with collision energy tuned at the stoponium mass. If this channel is kinematically closed stoponium could be discovered in $gg$, $γγ$ and $ZZ$ decay channels but higher statistics are needed. Effects of the stoponium-Higgs mixing and degeneracy are briefly discussed.
The existence of extra chiral generations with all fermions heavier than $M_Z$ is strongly disfavored by the precision electroweak data. The exclusion of one additional generation of heavy fermions in SUSY extension of Standard Model is less forbidden if chargino and neutralino have low degenerate masses with $Δm \simeq 1$ GeV. However the data are fitted nicely even by a few extra generations, if one allows neutral leptons to have masses close to 50 GeV. Such heavy neutrino can be searched in the reaction $e^+ e^- \to N\bar{N}γ$ at LEP-200 with total final luminosity of $2600 pb^{-1}$.
Using the computer code CompHEP we estimate the number of events and the background, at LEP II and TESLA, for the reaction $e^+ e^- \to N\bar{N}γ$, where $N$ is a hypothetical Dirac neutrino with mass of the order of 50 GeV.
A serious computational problem in the evaluation of hadronic collision processes is connected with the large number of partonic subprocesses included in the calculation. These are from the quark and gluon content of the initial hadrons, and from CKM quark mixing. For example, there are 180 subprocesses which contribute to the $W$+2jets process, and 292 subprocesses in $W$+3jets production at the LHC, even when quarks from only the first two generations are taken into account. We propose a simple modification of the rules for evaluation of cross sections and distributions, which avoids multiplication of channels from the mixture of quark states. The method is based on a unitary rotation of down quarks, thus, transporting the mixing matrix elements from vertices of Feynman diagrams to the parton distribution functions (PDF). As a result, one can calculate cross sections with significantly fewer subprocesses. For the example mentioned above, with the new rules, one need evaluate only 21 and 33 subprocesses respectively. The matrix elements of the subprocesses are calculated without quark mixing but with a modified PDF convolution which depends on the quark mixing angle, and on the topologies of gauge invariant classes of diagrams. The proposed method has been incorporated into the CompHEP program and checked with various examples.
The bounds on the anomalous contributions to the ggh and zgh vertices that can be obtained via the process eg->eh are discussed through a model independent analysis based on SU(2) X U(1) invariant operators of dim=6. A light Higgs boson with m_H ~ 120 GeV and center-of-mass energies of \sqrt{S}=500 GeV and \sqrt{S}=1500 GeV are assumed. The process eg->eh is shown to provide an excellent way to strongly constraint both the CP-even and CP-odd anomalous contributions in the ggH and ZgH couplings.
We consider the possibility of studying anomalous contributions to the gamma-gamma-H and Z-gamma-H vertices through the process e-gamma--> e-H at future e-gamma linear colliders, with Sqrt(S)=500-1500 GeV. We make a model independent analysis based on SU(2)xU(1) invariant effective operators of dim=6 added to the standard model lagrangian. We consider a light Higgs boson (mostly decaying in bar(b)-b pairs), and include all the relevant backgrounds. Initial e-beam polarization effects are also analyzed. We find that the process e-gamma--> e-H provides an excellent opportunity to strongly constrain both the CP-even and the CP-odd anomalous contributions to the gamma-gamma-H and Z-gamma-H vertices.
One-loop production of a Higgs boson in electro-photon collisions at future accelerators is studied via the process (e gamma -> e H), for intermediate Higgs masses. Exact cross sections, including the possibility of longitudinally polarized initial beams, are presented. Confirming previous estimates made in the Weizsäcker-Williams approximation, they are found to be more than two orders of magnitude larger than the cross sections for the crossed process (e e-> gamma H), in the energy range $\sqrt{s}=(0.5-2)$ TeV. We show that, not only (e gamma -> e H) has a similar potential as the (gamma gamma -> H) process for testing the one-loop gamma-gamma-H vertex, but, by requiring a final electron tagged at large angle in (e gamma -> e H), the (H e) production provides an excellent way of testing the Z-gamma-H vertex, too. Kinematical distributions for the (e gamma -> e H -> e bb) process with a tagged final electron are analyzed, and strategies for controlling the main irreducible background are found. Initial-state-radiation effects are checked to be within a few percent.
We study the bounds on the anomalous contributions to the gamma-gamma-H and Z-gamma-H vertices that can be obtained via the process e-gamma-->e H. We consider the representative case of an intermediate Higgs mass production of MH=120 GeV and for a center of mass energy of Sqrt{S}=500 GeV and Sqrt{S}=1500 GeV. We use a model independent analysis based on SU(2)xU(1) invariant operators of dimension 6 added to the Standard Model lagrangian. We find that this process provides an excelent way to put strong constraints both in the sector of CP-even and CP-odd anomalous couplings contribution to the gamma-gamma-H and Z-gamma-H vertices.
We discuss the SM Higgs discovery potential of the LHC in the channel pp->H+jet->gamma+gamma+jet when the jet is observed at sufficiently high Pt and a small rapidity to be reliably identified. We calculate all the signal subprocesses and the irreducible background with realistic kinematical cuts. The reducible QCD background is also estimated. We conclude that the channel gamma+gamma+jet can give about 120-200 signal events for Higgs mass MH=100-140 GeV at the integrated luminosity of 30 inverse fb. This signal rate should be compared with only 330-600 events for the irreducible background per two-photon invariant mass interval of 2 GeV. We estimate the QCD reducible background at the level of 20% of the irreducible one. Thus, one may hope that the Higgs boson can be discovered already during the LHC operation at a low luminosity. At a high luminosity the observation of several hundreds of high Pt Higgs bosons in this channel will be possible with significance higher than 15 for L=100 inverse fb.
Higgs production in $eγ$ collisions, through the one-loop reaction $eγ\to eH$ at large $p_T$, can provide a precise determination of the $ZγH$ vertex.
We show that the Higgs production in $eγ$ collisions, through the one-loop reaction $eγ\to eH$ at large transverse momentum, can provide a precise determination of the $ZγH$ vertex. We concentrate on the analysis of signal rates vs background looking for strategies to improve the ratio $S/B$. Possible bounds on anomalous contributions to the $ZγH$ vertex are derived.