-
Detecting Signs of Model Change with Continuous Model Selection Based on Descriptive Dimensionality
Authors:
Kenji Yamanishi,
So Hirai
Abstract:
We address the issue of detecting changes of models that lie behind a data stream. The model refers to an integer-valued structural information such as the number of free parameters in a parametric model. Specifically we are concerned with the problem of how we can detect signs of model changes earlier than they are actualized. To this end, we employ {\em continuous model selection} on the basis o…
▽ More
We address the issue of detecting changes of models that lie behind a data stream. The model refers to an integer-valued structural information such as the number of free parameters in a parametric model. Specifically we are concerned with the problem of how we can detect signs of model changes earlier than they are actualized. To this end, we employ {\em continuous model selection} on the basis of the notion of {\em descriptive dimensionality}~(Ddim). It is a real-valued model dimensionality, which is designed for quantifying the model dimensionality in the model transition period. Continuous model selection is to determine the real-valued model dimensionality in terms of Ddim from a given data. We propose a novel methodology for detecting signs of model changes by tracking the rise-up of Ddim in a data stream. We apply this methodology to detecting signs of changes of the number of clusters in a Gaussian mixture model and those of the order in an auto regression model. With synthetic and real data sets, we empirically demonstrate its effectiveness by showing that it is able to visualize well how rapidly model dimensionality moves in the transition period and to raise early warning signals of model changes earlier than they are detected with existing methods.
△ Less
Submitted 23 February, 2023;
originally announced February 2023.
-
Fluorine solubility and superconducting properties of Sm(O,F)BiS$_2$ single crystals
Authors:
Koki Kinami,
Yuji Hanada,
Masanori Nagao,
Akira Miura,
Shigeto Hirai,
Yuki Maruyama,
Satoshi Watauchi,
Isao Tanaka
Abstract:
Sm(O,F)BiS2 superconducting single crystals with various nominal F contents have been grown using KI-KCl flux. The solid solution limit of F at the O-site in Sm(O,F)BiS2 single crystals was approximately less than 20 at%. F concentrations of Sm(O,F)BiS2 single crystals were hardly controlled by nominal F contents. It suggests the existence of intrinsic stable phase. Superconductivity was shown at…
▽ More
Sm(O,F)BiS2 superconducting single crystals with various nominal F contents have been grown using KI-KCl flux. The solid solution limit of F at the O-site in Sm(O,F)BiS2 single crystals was approximately less than 20 at%. F concentrations of Sm(O,F)BiS2 single crystals were hardly controlled by nominal F contents. It suggests the existence of intrinsic stable phase. Superconductivity was shown at around 5 K. Superconducting transition temperature with zero resistivity were increased with increasing nominal F contents within less than 70 at% in O-site, despite similar F concentration and c-axis lattice constants.
△ Less
Submitted 17 June, 2021;
originally announced June 2021.
-
Spectral Pruning: Compressing Deep Neural Networks via Spectral Analysis and its Generalization Error
Authors:
Taiji Suzuki,
Hiroshi Abe,
Tomoya Murata,
Shingo Horiuchi,
Kotaro Ito,
Tokuma Wachi,
So Hirai,
Masatoshi Yukishima,
Tomoaki Nishimura
Abstract:
Compression techniques for deep neural network models are becoming very important for the efficient execution of high-performance deep learning systems on edge-computing devices. The concept of model compression is also important for analyzing the generalization error of deep learning, known as the compression-based error bound. However, there is still huge gap between a practically effective comp…
▽ More
Compression techniques for deep neural network models are becoming very important for the efficient execution of high-performance deep learning systems on edge-computing devices. The concept of model compression is also important for analyzing the generalization error of deep learning, known as the compression-based error bound. However, there is still huge gap between a practically effective compression method and its rigorous background of statistical learning theory. To resolve this issue, we develop a new theoretical framework for model compression and propose a new pruning method called {\it spectral pruning} based on this framework. We define the ``degrees of freedom'' to quantify the intrinsic dimensionality of a model by using the eigenvalue distribution of the covariance matrix across the internal nodes and show that the compression ability is essentially controlled by this quantity. Moreover, we present a sharp generalization error bound of the compressed model and characterize the bias--variance tradeoff induced by the compression procedure. We apply our method to several datasets to justify our theoretical analyses and show the superiority of the the proposed method.
△ Less
Submitted 13 July, 2020; v1 submitted 26 August, 2018;
originally announced August 2018.
-
Superconducting transition temperatures in the electronic and magnetic phase diagrams of Sr2VFeAsO3-delta, a superconductor
Authors:
Yujiro Tojo,
Taizo Shibuya,
Tetsuro Nakamura,
Koichiro Shoji,
Hirotaka Fujioka,
Masanori Matoba,
Shintaro Yasui,
Mitsuru Itoh,
Soshi Iimura,
Hidenori Hiramatsu,
Hideo Hosono,
Shigeto Hirai,
Wendy Mao,
Shinji Kitao,
Makoto Seto,
Yoichi Kamihara
Abstract:
We elucidate the magnetic phases and superconducting transition temperatures (Tc) in Sr2VFeAsO3-delta (21113V), an iron-based superconductor with a thick-blocking layer fabricated from a perovskite-related transition metal oxide. At low temperatures (T < 37.1 K), 21113V exhibited a superconducting phase in the range 0.031 =< delta =< 0.145 and an antiferromagnetic (AFM) iron sublattice in the rang…
▽ More
We elucidate the magnetic phases and superconducting transition temperatures (Tc) in Sr2VFeAsO3-delta (21113V), an iron-based superconductor with a thick-blocking layer fabricated from a perovskite-related transition metal oxide. At low temperatures (T < 37.1 K), 21113V exhibited a superconducting phase in the range 0.031 =< delta =< 0.145 and an antiferromagnetic (AFM) iron sublattice in the range 0.267 =< delta =< 0.664. Mixed-valent vanadium exhibited a dominant AFM phase in 0.031 =< delta =< 0.088, and a partial ferrimagnetic (Ferri.) phase in the range 0.124 =< delta =< 0.664. The Ferri. phase was the most dominant at a delta value of 0.267, showing an AFM phase of Fe at T < 20 K. Increasing the spontaneous magnetic moments reduced the magnetic shielding volume fraction due to the superconducting phase. This result was attributed to the magnetic phase of vanadium, which dominates the superconductivity of Fe in 21113V. The Tc-delta curve showed two maxima. The smaller and larger of Tc maxima occurred at delta = 0.073 and delta = 0.145, respectively; the latter resides on the phase boundary between AFM and the partial Ferri. phases of vanadium. 21113V is a useful platform for verifing new mechanisms of Tc enhancement in iron-based superconductors.
△ Less
Submitted 12 December, 2018; v1 submitted 12 February, 2018;
originally announced February 2018.
-
Upper Bound on Normalized Maximum Likelihood Codes for Gaussian Mixture Models
Authors:
So Hirai,
Kenji Yamanishi
Abstract:
This paper shows that the normalized maximum likelihood~(NML) code-length calculated in [1] is an upper bound on the NML code-length strictly calculated for the Gaussian Mixture Model. When we use this upper bound on the NML code-length, we must change the scale of the data sequence to satisfy the restricted domain. However, we also show that the algorithm for model selection is essentially univer…
▽ More
This paper shows that the normalized maximum likelihood~(NML) code-length calculated in [1] is an upper bound on the NML code-length strictly calculated for the Gaussian Mixture Model. When we use this upper bound on the NML code-length, we must change the scale of the data sequence to satisfy the restricted domain. However, we also show that the algorithm for model selection is essentially universal, regardless of the scale conversion of the data in Gaussian Mixture Models, and that, consequently, the experimental results in [1] can be used as they are. In addition to this, we correct the NML code-length in [1] for generalized logistic distributions.
△ Less
Submitted 18 November, 2018; v1 submitted 4 September, 2017;
originally announced September 2017.
-
Electronic structure of spin frustrated magnets: Mn$_3$O$_4$ spinel and postspinel
Authors:
S. Hirai,
Y. Goto,
A. Wakatsuki,
Y. Kamihara,
M. Matoba,
W. L. Mao
Abstract:
Mn$_3$O$_4$ is a spin frustrated magnet that adopts a tetragonally distorted spinel structure at ambient conditions and a CaMn$_2$O$_4$-type postspinel structure at high pressure. We conducted both optical measurements and \emph{ab} \emph{initio} calculations, and systematically studied the electronic band structures of both the spinel and postspinel Mn$_3$O$_4$ phases. For both phases, theoretica…
▽ More
Mn$_3$O$_4$ is a spin frustrated magnet that adopts a tetragonally distorted spinel structure at ambient conditions and a CaMn$_2$O$_4$-type postspinel structure at high pressure. We conducted both optical measurements and \emph{ab} \emph{initio} calculations, and systematically studied the electronic band structures of both the spinel and postspinel Mn$_3$O$_4$ phases. For both phases, theoretical electronic structures are consistent with the optical absorption spectra, and display characteristic band-splitting of the conduction band. The band gap obtained from the absorption spectra is 1.91(6) eV for the spinel phase, and 0.94(2) eV for the postspinel phase. Both phases are charge-transfer type insulators. The Mn 3\emph{d} $t_2$$_g$ and O 2\emph{p} form antibonding orbitals situated at the conduction band with higher energy.
△ Less
Submitted 12 June, 2014;
originally announced June 2014.
-
Giant atomic displacement induced by built-in strain in metastable Mn$_3$O$_4$
Authors:
Shigeto Hirai,
Antonio M. dos Santos,
Maxwell C. Shapiro,
Jamie J. Molaison,
Neelam Pradhan,
Malcolm Guthrie,
Christopher A. Tulk,
Ian R. Fisher,
Wendy L. Mao
Abstract:
We present x-ray, neutron scattering and heat capacity data that reveal a coupled first-order magnetic and structural phase transition of the metastable mixed-valence post-spinel compound Mn$_3$O$_4$ at 210 K. Powder neutron diffraction measurements reveal a magnetic structure in which Mn$^{3+}$ spins align antiferromagnetically along the edge-sharing \emph{a}-axis, with a magnetic propagation vec…
▽ More
We present x-ray, neutron scattering and heat capacity data that reveal a coupled first-order magnetic and structural phase transition of the metastable mixed-valence post-spinel compound Mn$_3$O$_4$ at 210 K. Powder neutron diffraction measurements reveal a magnetic structure in which Mn$^{3+}$ spins align antiferromagnetically along the edge-sharing \emph{a}-axis, with a magnetic propagation vector k = [1/2, 0, 0]. In contrast, the Mn$^{2+}$ spins, which are geometrically frustrated, do not order until a much lower temperature. Although the Mn$^{2+}$ spins do not directly participate in the magnetic phase transition at 210 K, structural refinements reveal a large atomic shift at this phase transition, corresponding to a physical motion of approximately 0.25 Å even though the crystal symmetry remains unchanged. This "giant" response is due to the coupled effect of built-in strain in the metastable post-spinel structure with the orbital realignment of the Mn$^{3+}$ ion.
△ Less
Submitted 8 November, 2012;
originally announced November 2012.
-
Normalized Maximum Likelihood Coding for Exponential Family with Its Applications to Optimal Clustering
Authors:
So Hirai,
Kenji Yamanishi
Abstract:
We are concerned with the issue of how to calculate the normalized maximum likelihood (NML) code-length. There is a problem that the normalization term of the NML code-length may diverge when it is continuous and unbounded and a straightforward computation of it is highly expensive when the data domain is finite . In previous works it has been investigated how to calculate the NML code-length for…
▽ More
We are concerned with the issue of how to calculate the normalized maximum likelihood (NML) code-length. There is a problem that the normalization term of the NML code-length may diverge when it is continuous and unbounded and a straightforward computation of it is highly expensive when the data domain is finite . In previous works it has been investigated how to calculate the NML code-length for specific types of distributions. We first propose a general method for computing the NML code-length for the exponential family. Then we specifically focus on Gaussian mixture model (GMM), and propose a new efficient method for computing the NML to them. We develop it by generalizing Rissanen's re-normalizing technique. Then we apply this method to the clustering issue, in which a clustering structure is modeled using a GMM, and the main task is to estimate the optimal number of clusters on the basis of the NML code-length. We demonstrate using artificial data sets the superiority of the NML-based clustering over other criteria such as AIC, BIC in terms of the data size required for high accuracy rate to be achieved.
△ Less
Submitted 16 May, 2012; v1 submitted 15 May, 2012;
originally announced May 2012.
-
Pressure-induced symmetry breaking in tetragonal CsAuI3
Authors:
Shibing Wang,
Shigeto Hirai,
Maxwell C. Shapiro,
Scott C. Riggs,
Theodore H. Geballe,
Wendy L. Mao,
Ian R. Fisher
Abstract:
Results of in situ high pressure x-ray powder diffraction on the mixed valence compound Cs2Au(I)Au(III)I6 (CsAuI3) are reported, for pressures up to 21 GPa in a diamond anvil cell under hydrostatic conditions. We find a reversible pressure-induced tetragonal to orthorhombic structural transition at 5.5-6 GPa, and reversible amorphization at 12-14 GPa. Two alternative structures are proposed for th…
▽ More
Results of in situ high pressure x-ray powder diffraction on the mixed valence compound Cs2Au(I)Au(III)I6 (CsAuI3) are reported, for pressures up to 21 GPa in a diamond anvil cell under hydrostatic conditions. We find a reversible pressure-induced tetragonal to orthorhombic structural transition at 5.5-6 GPa, and reversible amorphization at 12-14 GPa. Two alternative structures are proposed for the high-pressure orthorhombic phase, and are discussed in the context of a possible Au valence transition.
△ Less
Submitted 24 October, 2012; v1 submitted 4 May, 2012;
originally announced May 2012.
-
Non-Gaussianity and finite length inflation
Authors:
Shiro Hirai,
Tomoyuki Takami
Abstract:
In the present paper, certain inflation models are shown to have large non-Gaussianity in special cases. Namely, finite length inflation models with an effective higher derivative interaction, in which slow-roll inflation is adopted as inflation and a scalar-matter-dominated period or power inflation is adopted as pre-inflation, are considered. Using Holman and Tolley's formula of the nonlineari…
▽ More
In the present paper, certain inflation models are shown to have large non-Gaussianity in special cases. Namely, finite length inflation models with an effective higher derivative interaction, in which slow-roll inflation is adopted as inflation and a scalar-matter-dominated period or power inflation is adopted as pre-inflation, are considered. Using Holman and Tolley's formula of the nonlinearity parameter $f^\textrm{\tiny flattened}_\textrm{\tiny NL}$, we calculate the value of $f^\textrm{\tiny flattened}_\textrm{\tiny NL}$. A large value of $f^\textrm{\tiny flattened}_\textrm{\tiny NL}(f^\textrm{\tiny flattened}_\textrm{\tiny NL} > 100)$ can be obtained for all of the models considered herein when the length of inflation is 60-63 $e$-folds and $f_\textrm{\tiny NL}$ has strong dependence on the length of inflation. Interestingly, this length is similar to that for the case in which the suppression of the CMB angular power spectrum of $l=2$ was derived using the inflation models described in our previous papers.
△ Less
Submitted 23 April, 2010; v1 submitted 2 February, 2010;
originally announced February 2010.
-
Scale dependence of the power spectrum of the curvature perturbation determined using a numerical method in slow-roll inflation
Authors:
Shiro Hirai,
Tomoyuki Takami
Abstract:
The Taylor expansion method has been used to investigate the scale dependence of the power spectrum of the curvature perturbation. In the present study, an alternative numerical method is used to clarify the $k$ dependence. Although there is thought to be no large difference between these two methods, some differences arise among various inflation models. For example, at $k$ = 1 Mpc, there is a…
▽ More
The Taylor expansion method has been used to investigate the scale dependence of the power spectrum of the curvature perturbation. In the present study, an alternative numerical method is used to clarify the $k$ dependence. Although there is thought to be no large difference between these two methods, some differences arise among various inflation models. For example, at $k$ = 1 Mpc, there is a 1.4 % difference in the power spectrum, and with respect to the angular power spectrum, the difference of the value of $χ^2$ nearly 10 occur in new inflation. However, in hybrid inflation, these differences do not occur. The time dependence of the inflationary and cosmological parameters is investigated, and differences among inflation models are clarified.
△ Less
Submitted 5 August, 2009;
originally announced August 2009.
-
Time dependence of cosmological and inflationary parameters in slow-roll inflation
Authors:
Shiro Hirai,
Tomoyuki Takami
Abstract:
The dependence of cosmological and inflationary parameters on time during the last 60 e-folds in inflation is investigated using a slow-roll inflation model. The time dependence of the inflaton field is calculated for the case of chaotic inflation by a numerical method rather than the familiar approximations. It is found that the Hubble parameter and the spectral index decrease in the last 60 e-…
▽ More
The dependence of cosmological and inflationary parameters on time during the last 60 e-folds in inflation is investigated using a slow-roll inflation model. The time dependence of the inflaton field is calculated for the case of chaotic inflation by a numerical method rather than the familiar approximations. It is found that the Hubble parameter and the spectral index decrease in the last 60 e-folds. The dependence of the power spectrum of the curvature perturbation on the size perturbation k is calculated numerically, and it is shown that the overall constant value of the power spectrum thus determined differs from that given by the familiar method. However, the k-dependent spectra of the numerical calculation, the Bessel function approximation, and the familiar Taylor expansion are all found to be consistent.
△ Less
Submitted 14 December, 2008; v1 submitted 5 December, 2008;
originally announced December 2008.
-
Effect of initial condition of inflation on power and angular power spectra in finite slow-roll inflation
Authors:
Shiro Hirai,
Tomoyuki Takami
Abstract:
The effect of the initial condition of inflation on the power spectra of scalar and tensor perturbations is estimated assuming a slow-roll inflation model. By defining a more general initial state in inflation particular properties of the power spectrum such as oscillation can be revealed. The behavior of the power spectrum is shown to exhibit a step-like variation with respect to finite inflati…
▽ More
The effect of the initial condition of inflation on the power spectra of scalar and tensor perturbations is estimated assuming a slow-roll inflation model. By defining a more general initial state in inflation particular properties of the power spectrum such as oscillation can be revealed. The behavior of the power spectrum is shown to exhibit a step-like variation with respect to finite inflation length in cases of both radiation- and scalar matter-dominated pre-inflation. The power spectrum is shown to oscillate in the radiation-dominated case. The effects of such a power spectrum on the TT and TE power spectra are examined for three typical slow-roll inflation models; a small-field model, a large field model, and a hybrid model, considering both pre-inflation models. It is found that the discrepancies between WMAP3 data and the Lambda CDM model, such as suppression of the spectrum at l=2, may be explained to a certain extent by the finite length of inflation for inflation of close to 60 e-folds. The small-field inflation model with scalar matter-dominated pre-inflation provides the best fit to the WMAP3 data among the models considered. Relatively large changes in the angular TT power spectrum occur in response to small changes in inflation length in the radiation-dominated pre-inflation models, and half cases do not fit well with the observed data. This behavior is considered to be attributable to the oscillatory behavior of the power spectrum. The scalar matter-dominated case is thus preferred to the radiation-dominated case of pre-inflation, independent of the length of inflation.
△ Less
Submitted 12 October, 2007;
originally announced October 2007.
-
Angular Power Spectrum in Modular Invariant Inflation Model
Authors:
Mitsuo J. Hayashi,
Shiro Hirai,
Tomoyuki Takami,
Yusuke Okame,
Kenji Takagi,
Tomoki Watanabe
Abstract:
We propose a scalar potential of inflation, motivated by modular invariant supergravity, and compute the angular power spectra of the adiabatic density perturbations that result from this model. The potential consists of three scalar fields, S, Y and T, together with two free parameters. By fitting the parameters to cosmological data at the fixed point T=1, we find that the potential behaves lik…
▽ More
We propose a scalar potential of inflation, motivated by modular invariant supergravity, and compute the angular power spectra of the adiabatic density perturbations that result from this model. The potential consists of three scalar fields, S, Y and T, together with two free parameters. By fitting the parameters to cosmological data at the fixed point T=1, we find that the potential behaves like the single-field potential of S, which slowly rolls down along the minimized trajectory in Y. We further show that the inflation predictions corresponding to this potential provide a good fit to the recent three-year WMAP data, e.g. the spectral index n_s = 0.951. The TT and TE angular power spectra obtained from our model almost completely coincide with the corresponding results obtained from the \LambdaCDM model. We conclude that our model is considered to be an adequate theory of inflation that explains the present data, although the theoretical basis of this model should be further explicated.
△ Less
Submitted 28 November, 2006; v1 submitted 24 September, 2006;
originally announced September 2006.
-
Angular Power Spectrum and Dilatonic Inflation in Modular-Invariant Supergravity
Authors:
Mitsuo J. Hayashi,
Shiro Hirai,
Yusuke Okame,
Tomoki Watanabe
Abstract:
The angular power spectrum is investigated in the model of supergravity, incorporating the target-space duality and the non-perturbative gaugino condensation in the hidden sector. The inflation and supersymmetry breaking occur at once by the interplay between the dilaton field as inflaton and the condensate gauge-singlet field. The model satisfies the slow-roll condition which solves the η-probl…
▽ More
The angular power spectrum is investigated in the model of supergravity, incorporating the target-space duality and the non-perturbative gaugino condensation in the hidden sector. The inflation and supersymmetry breaking occur at once by the interplay between the dilaton field as inflaton and the condensate gauge-singlet field. The model satisfies the slow-roll condition which solves the η-problem. When the particle rolls down along the minimized trajectory of the potential at a duality invariant fixed point T=1, we can obtain the e-fold value \sim 57. And then the cosmological parameters obtained from our model well match with the recent WMAP data combined with other experiments. The TT and TE angular power spectra also show that our model is compatible with the data for l > 20. However, the best fit value of τin our model is smaller than that of the ΛCDM model. These results suggest that, among supergravity models of inflation, the modular-invariant supergravity seems to open a hope to construct the realistic theory of particles and cosmology.
△ Less
Submitted 23 January, 2006;
originally announced January 2006.
-
Effect of the length of inflation on angular TT and TE power spectra in power-law inflation
Authors:
Shiro Hirai,
Tomoyuki Takami
Abstract:
The effect of the length of inflation on the power spectra of scalar and tensor perturbations is estimated using the power-law inflation model with a scale factor of a(t) = t^q. Considering various pre-inflation models with radiation-dominated or scalar matter-dominated periods before inflation in combination with two matching conditions, the temperature angular power spectrum (TT) and temperatu…
▽ More
The effect of the length of inflation on the power spectra of scalar and tensor perturbations is estimated using the power-law inflation model with a scale factor of a(t) = t^q. Considering various pre-inflation models with radiation-dominated or scalar matter-dominated periods before inflation in combination with two matching conditions, the temperature angular power spectrum (TT) and temperature-polarization cross-power spectrum (TE) are calculated and a likelihood analysis is performed. It is shown that the discrepancies between the Wilkinson Microwave Anisotropy Probe (WMAP) data and the LCDM model, such as suppression of the spectrum at l = 2,3 and oscillatory behavior, may be explained by the finite length of inflation model if the length of inflation is near 60 e-folds and q > 300. The proposed models retain similar values of chi^2 to that achieved by the LCDM model with respect to fit to the WMAP data, but display different characteristics of the angular TE power spectra at l < 20.
△ Less
Submitted 13 December, 2005;
originally announced December 2005.
-
Length of inflation and WMAP data in the case of power-law inflation
Authors:
Shiro Hirai,
Tomoyuki Takami
Abstract:
The effect of the length of inflation on the power spectra of scalar and tensor perturbations is estimated using the power-law inflation model with a scale factor of a(t) = t^q. Considering various pre-inflation models with radiation-dominated or matter-dominated periods before inflation in combination with two matching conditions, the power spectrum of curvature perturbations at large scales is…
▽ More
The effect of the length of inflation on the power spectra of scalar and tensor perturbations is estimated using the power-law inflation model with a scale factor of a(t) = t^q. Considering various pre-inflation models with radiation-dominated or matter-dominated periods before inflation in combination with two matching conditions, the power spectrum of curvature perturbations at large scales is calculated. Comparison of the derived angular power spectrum with the Wilkinson Microwave Anisotropy Probe (WMAP) data reveals a possibility that the WMAP can be explained by the finite length of inflation model if the length of inflation is near 60 e-folds and q>200.
△ Less
Submitted 21 June, 2005;
originally announced June 2005.
-
Pre-inflation models and WMAP data
Authors:
Shiro Hirai
Abstract:
The effect of pre-inflation physics on the power spectrum of scalar perturbations is estimated and there is a possibility that the pre-inflation physics explains the Wilkinson Microwave Anisotropy Probe data if the length of inflation is near 60 e-folds. Considering various pre-inflation models with radiation-dominated or matter-dominated periods before inflation, the power spectrum of curvature…
▽ More
The effect of pre-inflation physics on the power spectrum of scalar perturbations is estimated and there is a possibility that the pre-inflation physics explains the Wilkinson Microwave Anisotropy Probe data if the length of inflation is near 60 e-folds. Considering various pre-inflation models with radiation-dominated or matter-dominated periods before inflation, the power spectrum of curvature perturbations for large scales is calculated, and the spectral index and running spectral index are derived.
△ Less
Submitted 27 April, 2004;
originally announced April 2004.
-
Effect of Pre-inflation Conditions on Scalar and Tensor Perturbations in Inflation
Authors:
Shiro Hirai
Abstract:
The effect of the initial conditions in inflation on scalar and tensor perturbations is investigated. Formulae for the power spectra of gravitational waves and curvature perturbations for any initial conditions in inflation are derived, and the ratio of scalar to tensor perturbations and spectral index are calculated. The formulation is applied to some simplified pre-inflationary cosmological mo…
▽ More
The effect of the initial conditions in inflation on scalar and tensor perturbations is investigated. Formulae for the power spectra of gravitational waves and curvature perturbations for any initial conditions in inflation are derived, and the ratio of scalar to tensor perturbations and spectral index are calculated. The formulation is applied to some simplified pre-inflationary cosmological models, and the differences of the ratio and spectral index are investigated with respect to two matching conditions. In addition, the present power spectrum of gravitational waves is derived. The proposed formulation is preliminarily shown to be a possible test of the appropriateness of a given pre-inflationary model.
△ Less
Submitted 24 July, 2003;
originally announced July 2003.
-
Initial condition of scalar perturbation in inflation
Authors:
Shiro Hirai
Abstract:
A formula for the power spectrum of curvature perturbations having any initial conditions in inflation is obtained. Based on the physical conditions before inflation, the possibility exists that the initial state of scalar perturbations is not only the Bunch-Davies state, but also a more general state (a squeezed state). For example, the derived formula for the power spectrum is calculated using…
▽ More
A formula for the power spectrum of curvature perturbations having any initial conditions in inflation is obtained. Based on the physical conditions before inflation, the possibility exists that the initial state of scalar perturbations is not only the Bunch-Davies state, but also a more general state (a squeezed state). For example, the derived formula for the power spectrum is calculated using simple toy cosmological models. When there exists a radiation-dominated period before inflation, the behavior of the scalar perturbation is revealed not to vary greatly; however, from large scales to small scales the power spectrum of the curvature perturbations oscillates around the normal value. In addition, when inflation has a large break and the breaking time is a radiation- dominated period, a large enhancement is revealed to occur which depends on the length of the breaking time.
△ Less
Submitted 4 December, 2002;
originally announced December 2002.
