Coming Seminars
Seminars in July
Speaker
Zhenjie Liu
Date/Place
13:30-, 2(Wed), June. @ES606
Title
Lyman-α Emitters for HSC-Medium Bands and Cosmological Forecast
Abstract
LAEs are promising tracers of large-scale structure at z > 2.5. Upcoming
wide-field imaging surveys using medium-band (MB) filters on the Subaru
Telescope (HSC) and Blanco Telescope (DECam) aim to detect LAEs, with
DESI-2 providing spectroscopic follow-up. To support the target selection
and scientific planning of the HSC-MB survey, we construct realistic mock
LAE catalogs and forecast the resulting cosmological constraints.
Seminars in June
Speaker
Nobufusa Kobayashi
Date/Place
11:00-, 25(Wed), June. @ES606
Title
Domain-wall quintessence (Progress report)
Abstract
Understanding the origin of dark energy, which drives the accelerated expansion of the universe, remains one of the central challenges in modern cosmology. In recent years, extensive efforts have been devoted to addressing this problem. Among the proposed candidates, quintessence models have received considerable attention. Moreover, observational evidence suggesting the presence of statistical anisotropy on cosmological scales has been gradually accumulating[1]. Given this context, there is a need for a theoretical framework that can simultaneously account for both quintessence and large-scale anisotropy.
A previous study [2] introduced phase defects a Hubble scale monopole into a quintessence model to reproduce an anisotropic universe. In the present work, we extend this approach by considering a domain wall as the phase defects. Specifically, we construct a novel theoretical model in which a flat domain wall with a thickness on the order of the Hubble scale generates a potential responsible for the currently observed accelerated expansion of the universe. In this model, it is analytically shown that the spacetime near the domain wall acquires an approximately homogeneous but anisotropic structure, which in turn induces global anisotropy in the universe. Furthermore, we demonstrate that the time evolution of this system depends solely on the vacuum expectation value of the scalar field, denoted by \eta, and through numerical analysis, we confirm that local accelerated expansion occurs under the condition \eta>>0.25 m_p. We also analyze the behavior of light along null geodesics in the constructed background spacetime, and by comparing the resulting distance moduli of Type Ia supernovae with observational data, we derive constraints on the spatial position of the domain wall in our universe.
[1] PERIVOLAROPOULOS, Leandros. arXiv:1401.5044v1
[2] SANCHEZ, Juan C. Bueno; PERIVOLAROPOULOS, Leandros. arXiv:1110.2587v3
Speaker
Akira Ishikawa
Date/Place
13:30-, 25(Wed), June. @ES606
Title
A Unified Analysis of Interacting Dark Energy Models Using IDECAMB
Abstract
I will give an interim report in preparation for my presentation at the summer school. In this talk, I will summarize my current understanding of the paper I am reviewing and outline my plans going forward.
In the current standard cosmological model, dark energy is considered to be the cosmological constant Λ. However, the results from the latest large-scale galaxy survey, DESI DR2, suggest deviations from the standard model [1].
One proposed model that could potentially explain such deviations is the Interacting Dark Energy (IDE) model, which assumes an interaction between dark energy and dark matter—two components believed to play different roles in the evolution of the universe. The IDE framework is expected to address several unresolved issues in cosmology. Indeed, attempts to explain the DESI observations using IDE models have been reported in [2] and [3].
While numerous IDE models have been proposed, comparing them with observational data has required individually modifying the CAMB package (a code used to compute CMB anisotropies) for each model. To address this issue, a modified version of CAMB, known as IDECAMB, has recently been developed and released [4]. IDECAMB allows systematic parameterization of models, enabling efficient analysis across a wide range of IDE models—including coupled quintessence models and coupled fluid models, which previously faced challenges such as large-scale instabilities. This advancement is expected to further accelerate progress in IDE research.
In this talk, we will present a comprehensive discussion on the current status and future prospects of IDE model studies, incorporating the latest observational results, through a review of reference [4].
[1] DESI Collaboration(2025), arXiv:2503.14738
[2] A. Chakraborty et al., arXiv:2503.10806
[3] W. Giarè et al., Phys. Rev. Lett, 133, 251003, 2024
[4] Yun-He Li and Xin Zhang, JCAP09(2023)046
Speaker
Ryota Himeno
Date/Place
13:30-, 25(Wed), June. @ES606
Title
Reconstructing the Matter Density Field at Recombination Using Machine Learning
Abstract
The current structure of galaxies (on scales of approximately 0.01 Mpc) has evolved through nonlinear growth influenced by gravity and other forces. As a result, the matter distribution observed in the present universe partially obscures the imprints of physical phenomena from the early universe. Due to this, it is currently difficult to discuss early-universe physics based solely on the observed present-day matter distribution. To address this issue, research has been conducted on reconstructing the matter distribution of the early universe from current matter distribution maps. Previous studies have developed algorithms to reconstruct baryon acoustic oscillations (BAO)—traces of sound waves generated in the early universe—based on linear perturbation theory [2]. However, it is known that reconstructions based on linear perturbation theory are less accurate on small scales, such as collapsed galaxies, where nonlinear gravitational evolution is significant.
In response, Shallue & Eisenstein (2023) introduced a machine learning-based approach in addition to traditional methods rooted in linear perturbation theory. Their method enables more accurate reconstruction of the early-universe matter distribution at small scales, such as halos and galaxies. Specifically, instead of explicitly modeling gravitational dynamics, their approach learns a direct mapping between the initial and final states in simulations, thereby achieving a more precise reconstruction of the initial density field. As a result, the scale at which the correlation between the initial and reconstructed power spectra exceeds 95% was improved by a factor of two.
This presentation reviews the study by Shallue & Eisenstein (2023) and discusses how their findings can be applied to the analysis of the current galaxy distribution.
[1] Shallue, Christopher J. & Eisenstein, Daniel J. (2023). Reconstructing cosmological initial conditions from late-time structure with convolutional neural networks, MNRAS.
[2] Eisenstein et al. (2007). Improving cosmological distance measurements by reconstruction of the baryon acoustic peak, The American Astronomical Society.
Speaker
Taiki Makino
Date/Place
13:30-, 18(Wed), June. @ES606
Title
A theoretical analysis of the spectrum of primordial gravitational waves based on the Standard Model
Abstract
TBA
Speaker
Fuyuko Tanaka
Date/Place
13:30-, 18(Wed), June. @ES606
Title
Mass estimation of nearby galaxy clusters using gravitational lensing effects
Abstract
The density fluctuations in the early universe grew amid the interplay between gravitational interactions caused by dark matter and accelerated expansion caused by dark energy, leading to the formation of celestial bodies such as galaxies and galaxy clusters. Galaxy clusters, in particular, are the most massive objects in the universe, composed mainly of hot gas and dark matter. Their total mass can be measured using the “weak gravitational lensing effect,” in which the gravity of the foreground galaxy cluster bends light from background galaxies. The mass function, which represents the number density of galaxy clusters as a function of mass, is sensitive to cosmological parameters such as the energy density of matter Ωm and the amplitude of matter density fluctuations σ8. Generally, galaxy clusters are detected using X-ray luminosity, Sunyaev-Zel'dovich signals, and the spatial concentration of galaxies. Establishing a mass-observational quantity relationship is essential to link the mass function with these observational quantities. In this study, we focus on X-ray-detected X-ray-luminous nearby galaxy clusters and aim to establish a precise mass-X-ray luminosity relationship. As a first step, we performed mass estimation using weak gravitational lensing effects for 9 galaxy clusters with high X-ray luminosities (0.1–2.4 keV) observed by the Subaru Telescope Hyper Suprime-Cam (HSC) at redshifts 0.03 < z < 0.12. We report the results of this study.
Speaker
Kohki Tanida
Date/Place
13:30-, 11(Wed), June. @ES606
Title
Toward a phenomenological test of gravity using the COCOA pipeline
Abstract
The current standard cosmological model, the ΛCDM model, describes the geometric and structural evolution of the universe and has been supported by various cosmological observations, including large-scale structures.
However, the ΛCDM model requires the assumption of the existence of unknown matter (dark matter) that serves as a gravitational source for galaxy formation, and unknown energy (cosmological constant or dark energy) that drives accelerated expansion. In the ΛCDM model, gravitational interactions are described by general relativity (GR), but attempts have also been made to explain accelerated expansion without introducing dark energy by modifying the theory of gravity at cosmological scales from GR. This is known as “modified gravity theory.” In recent years, improvements in observational accuracy have yielded results from large-scale structure measurements that suggest a breakdown of the ΛCDM model. It is expected that future surveys, such as the Nancy Grace Roman Space Telescope (hereinafter Roman), will further advance the verification of modified gravity theory on a cosmological scale.
We have implemented the Boltzmann code MGCAMB, which enables power spectrum calculations compatible with modified gravity theory, into the cosmological analysis pipeline COCOA currently under development for Roman. Furthermore, we aim to analyze the final data from the Subaru Telescope HSC in advance and verify gravity theory. MGCAMB is independent of specific theories and can handle modifications to gravity felt by non-relativistic/relativistic components as phenomenological parameters (μ-Σ parameters).
In this presentation, we will discuss the predicted results of cosmological parameter constraints using COCOA with MGCAMB and the prospects for constraints on modified gravity theories.
Speaker
Keitaro Ishikawa
Date/Place
13:30-, 4(Wed), June. @ES606
Title
Differentiable Cosmological Emulators for Halo Assembly Bias Studies
Abstract
Galaxies form in dark matter haloes. The spatial distribution of dark matter haloes, and the distribution and the number of galaxies within a dark matter halo, depend primarily on the halo mass. However, they are also known to depend on halo properties other than mass, such as halo formation history (Wechsler et al. 2006). This secondary dependence is called assembly bias. In this work, to construct a halo statistics emulator that also predicts the assembly bias, we focused on the concentration of haloesas a representative secondary parameter, and measured the cross-correlation function of various halo samples selected according to the mass and concentration using Dark Quest II simulation data. We then constructed an accurate power spectrum emulator for these statistics as inputs of halo mass and concentration using a feed-forward neural network. By taking partial derivatives of the emulator output with respect to halo mass, we
demonstrate that the network successfully captures how variations in concentration modulate the power amplitude. In this talk, we will discuss the implementation of automatic differentiation (AD) for both emulator-based gradient evaluations and point‐estimation workflows,highlighting the accuracy advantages of AD based on JAX or PyTorch over simple finite‐difference approximations when compared to simulated galaxy power spectra. Finally, we will outline prospects for integrating automatic‐differentiation‐compatible Hamiltonian Monte Carlo (HMC) methods into cosmological inference pipelines, exploiting the synergistic strengths of AD and HMC for efficient, high‐dimensional parameter exploratio
Seminars in May
Speaker
Shohei Saga
Date/Place
13:30-, 28(Wed), May. @ES606
Title
Relativistic effects on redshift-space distortions: towards
detection with Euclid
Abstract
The observed galaxy distribution via galaxy redshift surveys appears distorted due to redshift-space distortions (RSD). One dominant contribution to RSD comes from the Doppler effect induced by the peculiar velocity of galaxies. The other relativistic effects induce the asymmetric distortions, which contains qualitatively different cosmological information from the standard RSD analysis. I will present my recent contributions to Euclid on the asymmetric galaxy clustering.
Speaker
Kiyotomo Ichiki
Date/Place
13:30-, 14(Wed), May. @ES606
Title
What does Planck tell us about inflation? - and beyond -
Abstract
The scattering of cosmic microwave background radiation (CMB) in galaxy clusters induces a polarization signal according to the quadrupole anisotropy of the photon distribution at the location of the cluster. This "remote quadrupole" derived from measurements of the induced polarization provides an opportunity to reconstruct primordial fluctuations on large scales. By comparing the local quadrupole of the CMB predicted by the primordial fluctuations reconstructed in this way with direct observations by CMB satellites, we discuss how the dark energy can be tested by CMB observations beyond the indefinite nature of the cosmic variance.
Speaker
Shuichiro Yokoyama
Date/Place
13:30-, 7(Wed), May. @ES606
Title
What does Planck tell us about inflation? - and beyond -
Abstract
Recent precise CMB measurements, such as Planck and BICEP/Keck, have given us a deeper understanding of cosmic inflation.In fact, conventional simple models (e.g., proposed in the early '80s) are in tension. In this talk, I would like to give a brief review of inflation, in particular, introduce how to classify the slow-roll inflationary models on the ns-r plane. I would also like to talk with you about what to expect in the future.
Seminars in April
Speaker
Shun Arai
Date/Place
13:30-, 30(Wed), Apr. @ES606
Abstract
Gravitational lensing has been a remarkable observational tool for over the
past century, providing insights into gravitational physics through its
relativistic and geometric signatures. Meanwhile, the large-scale structure
(LSS) of the Universe encapsulates the dynamic interplay between
gravitational infall and cosmic expansion. Recent literature of cosmology
has drawn the attention to utilise gravitational lensing of the cosmic
microwave background (CMB lensing), sourced by LSS, to reveal the detailed
evolutionary history of cosmic structures, especially during their early
formation stages at high redshifts (z > 1).
In this seminar, we will introduce the concept of "CMB-lensing tomography,"
a method that employs angular two-point correlation statistics measured
across distinct redshift slices. We will outline a strategic programme
aimed at achieving precise CMB-lensing tomography at high redshifts. To the
last minutes, we will illustrate this methodology using recent observations
of Lyman-break galaxies obtained with the Subaru Telescope.
Speaker
Akira Ishikawa
Date/Place
13:30-, 23(Wed), Apr. @ES606
Title
Self Introduction
Abstract
Self Introduction
Speaker
Huyuko Tanaka
Date/Place
13:30-, 23(Wed), Apr. @ES606
Title
Self Introduction
Abstract
Self Introduction
Speaker
Ryota Himeno
Date/Place
13:30-, 23(Wed), Apr. @ES606
Title
Self Introduction
Abstract
Self Introduction
Speaker
Taisuke Makino
Date/Place
13:30-, 16(Wed), Apr. @ES606
Title
Self Introduction
Abstract
Self Introduction
Speaker
Nobuhusa Kobayashi
Date/Place
13:30-, 16(Wed), Apr. @ES606
Title
Empirical Galaxy Formation Modeling
Abstract
Self introduction
Speaker
Kiyotomo Ichiki
Date/Place
13:30-, 16(Wed), Apr. @ES606
Title
Policy speech
Abstract
Policy speech