Recovering the initial conditions of our local Universe from NOG and PSCz catalogues

We apply the ztrace algorithm to the optical NOG and infrared PSC z galaxy catalogues to reconstruct the pattern of primordial fluctuations that have generated our local Universe. We check that the density fields traced by the two catalogues are well correlated, and consistent with a linear relation [either in δ or in  log (1 +δ)  ] with relative bias (of NOG with respect to PSC z )   b _rel= 1.1 ± 0.1  . The relative bias relation is used to fill the optical zone of avoidance at  | b | < 20°  using the PSC z galaxy density field.
We perform extensive testing on simulated galaxy catalogues to optimize the reconstruction. The quality of the reconstruction is predicted to be good at large scales, up to a limiting wavenumber   k _lim≃ 0.4 h Mpc⁻¹  beyond which all information is lost. We find that the improvement arising from the denser sampling of the optical catalogue is compensated by the uncertainties connected to the larger zone of avoidance.
The initial conditions reconstructed from the NOG catalogue are found (analogously to those from the PSC z ) to be consistent with a Gaussian paradigm. We use the reconstructions to produce sets of initial conditions ready to be used for constrained simulations of our local Universe.     

Cosmological constraints from Chandra X-ray observations of galaxy clusters

Chandra X-ray observations of rich, dynamically relaxed galaxy clusters allow the properties of the X-ray gas and the total gravitating mass to be determined precisely. Here, we discuss how Chandra observations may be used as a powerful tool for cosmological studies. By combining Chandra X-ray results on the X-ray gas mass fractions in clusters with independent measurements of the Hubble constant and the mean baryonic matter density of the universe, we obtain a tight constraint on the mean total matter density of the universe, Ο_m, and an interesting constraint on the cosmological constant, Ο_Λ. Using these results, together with the observed local X-ray luminosity function of the most X-ray luminous galaxy clusters, a mass-luminosity relation determined from Chandra and ROSAT X-ray data and weak gravitational lensing observations, and the mass function predicted by numerical simulations, we obtain a precise constraint on the normalization of the power spectrum of density fluctuations in the nearby universe,σ₈. We compare our results with those obtained from other, independent methods. This revised version was published online in July 2006 with corrections to the Cover Date.     

动态最小二乘支持向量机学习算法

针对大样本集的训练问题和动态训练样本的模型更新问题,提出了动态最小二乘支持向量机学习算法.该算法充分利用已建好的模型,逐渐加入新样本,并可删除位于任何位置的非支持向量,避免了矩阵求逆运算,保证了算法的高效率.大坝变形及电离层延迟两个时间序列的预报实例表明,该算法具有计算时间短、预报精度高的特点.