Project Expo 2024

13:30–15:00, Friday, May 10th, 2024

Lecture Hall, Kavli IPMU

CD3 Year 1 Summary

Jia Liu

Dark matter detection at COSI �(March 21-22, 2024)

Future Science with CMB x LSS (April 10-14, 2023, YITP, Kyoto)

AI-driven discovery in physics and �astrophysics (January 22-26, 2024)

Baryons in the Universe 2024 (April 8-12, 2024)

CD3 x CMB x Astro Seminar by Jo Dunkley �(Feb. 5, 2024)

Housekeeping & Announcements

• “CD3 x astro/math/pheno/CMB/neutrino” seminars (Contact: Leander)
• Your speaker may be funded by CD3 if they work on data science/AI/ML
• Hack Fridays, 13:30-17pm (Contacts: Cesar, Boris, Linda, Jingjing)
• Playground for data-related ideas with presentations, study groups, and project reports
• Currently–summer: study group on AI/ML basics
• Format constantly evolving, input welcome!
• Joining CD3: sign up on https://cd3.ipmu.jp/ (Contact: Katya)
• Stay up to date with the future
• Funding for workshops, visitors, and travel
• Transition to non-academic career
• Required commitments: engage in CD3 activities, publish with CD3 affiliation, follow the Code of Conduct

Code of Conduct

CD3 is committed to creating a safe, inclusive, and respectful environment for all members of our community. CD3 members are required to:

Treat each other with respect and professionalism. Discrimination, harassment, and retaliation of any kind will not be tolerated.

Follow the highest standards of academic integrity and honesty. This includes, but is not limited to, avoiding plagiarism, fabrication, and falsification of data. All members are required to comply with Kavli IPMU’s Research Ethics.

Researchers using AI tools should document this use in the methods, acknowledgements, or other appropriate sections in the paper. Do not use AI tools as a credited author on a research paper, as it can not carry accountability.

Moving Forward

• Science: brainstorm key projects and form interdisciplinary teams
• In planning (late June): project brainstorm led by CD3 senior members (coming up next)�––> Goal: formation of interdisciplinary research teams on identified key projects
• Long-term: data and simulation center for IPMU led-projects�––> Expect us to ask about your computing needs soon
• Education: build a network in Asia
• AstroAI Asian Network (A³ Net, https://cd3.ipmu.jp/a3n/)�––> Goal: Connecting 10+ Asian institutes on Astro x AI education and future collaborations, with yearly summer schools
• First summer school in Osaka, Sept 2-6, 2024

Incoming Postdocs

New Faculty Members

Special thanks!

Sherry Song, Tianyu Zhu, Daniela Galarraga-Espinosa, Jiaxi Yu, Guillermo Pascual Cisneros, Minh Nhat Nguyen

Jingjing Shi, Linda Blot, Leander Thiele

CD3 Project Expo 2024

15 projects x 5 minutes. Introduced by non-experts�Stay after for a group photo!

Tom Melia

Expert: Mark Vagins

Underground neutrino observatory

Science: detect and study solar and atmospheric neutrinos. Also search for proton decay

Water (50,000 tons) Cherenkov detector. Neutrino scatter off electrons or nucleus, relativistic charged particle detected through Cherenkov light cone, picked up in photomultipliers

Tom Melia

Expert: Mark Vagins

Gadolinium upgrade (SuperK-Gd)

Science: detect relic supernova neutrinos

Trigger rate (20kHz event candidates at 3.5 MeV) goes up an order of magnitude for each MeV the energy threshold is lowered. Real time parallel online computing

Data is ~ PMT hits with GPS time stamps

Magnetic tape storage 14 Tb/yr raw data

Extensive MC simulation

Hiromi Yokoyama

Expert: César Jesús-Valls

• Precision neutrino oscillation measurements.
• Leading experiment regarding searches for CP-violation (comparing neutrino vs antineutrino oscillations).

Science goals

Shaping current knowledge of neutrino oscillations.

Working principle:

1) Produce accelerator neutrinos.

2) Measure neutrino beam before it oscillates with near detectors.

3) Measure neutrino beam after neutrino oscillations occur with SK.

4) Compare near and far detector data to measure oscillations!

Taking data since 2010.

> 500 collaborators!!

Hiromi Yokoyama

Expert: César Jesús-Valls

Synergies with CD3

• Event reconstruction (going from detector hits to event information) presents unique challenges in each detector.
• Statistical analysis of the T2K’s huge datasets, crucial to expand understanding of neutrino oscillations.

Detector roles:

​

Super-Kamiokande: Far from the accelerator we have few neutrinos, we need huge detector mass!

ND280: Near from accelerator we have more data, but we need great precision to control systematics!

Near Detectors are completely different to Super-Kamiokande!!

A drawing of the ND280 detector

(recently upgraded!)

This is how neutrinos look like in ND280’s data!

T2K CM last summer in Ibaraki!

flagship exp.!

Linda Blot

Expert: Masaki Yamashita

Science Goal:

• Direct detection of Dark Matter (DM) via elastic scattering off nuclei
• Measuring mass and cross section of DM particle

Rare event:

• Observing from the underground laboratory in Italy (LNGS), 1400 m depth
• Data taking was started from 2021
• More than 5 years operation is planned

​

Looking for dark matter from underground

Detector:

Gas-liquid dual phase Xe Time projection chamber (TPC)

• 8.5 tonne Xenon
• ~500 Photosensor(PMT) (Top and Bottom array)
• Immersed in the gadolinium-loaded water

Gd-loaded Water Tank

LXe Detector

Photosensor(PMT) array

Linda Blot

Expert: Masaki Yamashita

XENON with CD3

• Data Analysis : Position reconstruction of events and waveform analysis by using Machine Learning
• Rare event search -> Data driven New Physics Search

Signal:

LXe scintillation light (S1) and ionization (S2)

=> energy, position, particle identification

S1

S2

Baptiste Jost

Expert: Takeo Higuchi

Overview:

• Belle II is an experiment on SuperKEKB
• SuperKEKB: accelerator in Tsukuba (3km circ)
• Electron/Positron accelerator (7 and 4 Gev)
• Data taking in 2019-2022 and back since February 2024

​

Science goals:

• Probe beyond standard model physics
• CP violation in the physics of B mesons: study

Baptiste Jost

Expert: Takeo Higuchi

19 detector modules produced in IPMU’s clean room, delivered in 2019

Source: arXiv:2402.17260

Data analysis challenge: flavor tagging

i.e. identify from from their disintegration products.

​

GFlaT: new machine learning tool using graph neural network.

Drastic improvement: 18% better tagging efficiency compared to previous algorithm (see arXiv:2402.17260)

Elisa Ferreira

Expert: Masahito Yamazaki

Overview:

• Quantum computing: uses specialized technology, including computer hardware and algorithms that take advantage of quantum mechanics, to solve complex problems that classical computers or supercomputers can't solve, or can't solve quickly enough.
• Quantum algorithm: is an algorithm that runs on a realistic model of quantum computation
• Quantum machine learning is the integration of quantum algorithms within machine learning programs

​

​

University of Tokyo and IBM's 127-qubit IBM Quantum Eagle processor. First utility-scale processor outside North-America

Elisa Ferreira

Expert: Masahito Yamazaki

Tensorflow quantum

Goal: The goal of the project is to study if/whether/how quantum algorithms will be of help in analyzing data or more generally solving computational problems, at least in the future where quantum computers will be computationally more powerful.

Ex.: - Tensorflow quantum (TFQ): TFQ allows researchers to construct quantum datasets, quantum models, both using quantum data and hybrid quantum-classical models.

• Project idea: use quantum algorithms to simulate fuzzy dark matter (described by a Schrodinger - Poisson system)

John Silverman

Expert: Hiromi Yokoyama

Science and Ethics Project

​

CD3:

Leander Thiele

(as Analysis supervisor)

Kanmi Nose (Masa’s Group)

Hiromi M. Yokoyama

​

​

​

John Silverman

Expert: Hiromi Yokoyama

Measuring and analyzing people's attitudes for science and ethics

​

Data-driven social science is new

​

César Jesús-Valls

Expert: John Silverman

Quasar host galaxies

(z ~ 0.1)

Merging black holes

(z ~ 0.5)

First massive black holes

(z > 7)

Primordial black holes

(z ~ 0)

Niikura, Takada et al. 2019

Thorne et al. in prep

Tang et al. 2021

Matsuoka, Onoue et al. 2019

https://pfs.ipmu.jp/instrumentation.html

Prime Focus Spectrograph (PFS) is a spectrograph system on Subaru telescope.

​

• What does it do?
• We can obtain the spectrum of an individual target on the sky..
• How does it work?
• Positioning the 2394 fibers to each target at the focal plane (FOV 1.25 deg²), and measure the spectrum of each for 380-1260 nm.
• When does it operate?
• The engineering observations have started in fall 2021 and in progress.
• The science observation is expected from Feb. 2025 after the verification of the performance.
• Why do we need this?
• see next page...

Follow the photons from the sky!

• 8.2m primary mirror: form a prime focus
• Wide Field Corrector: ensure to have sufficient focal plane area
• Cobra in Prime focus instrument: the photons from each sky position are collected in each fiber
• Spectrograph: the photons in each fiber will be expanded spatially in color by gratings
• The photons will be converted to the electric signal via the CCD arrays.

Tomotake Matsumura

Expert: KG Lee

Tomotake Matsumura

Expert: KG Lee

The science goals of PFS on Subaru are

• cosmology
• galaxy evolution
• galaxy archaeology
• and more

• Guaranteed signal Measurements of the total neutrino mass to the precision of
• addressing the neutrino mass hierarchy
• Discovery potential End-to-end test of the Universe: Hubble tension, time evolving dark energy, test GR, curvature of space, primordial power spectrum, and more

From 2D to 3D!

redshift z of 2.4 ~ 0.6

cosmology

First Public Data Released by the Hyper Suprime-Cam Subaru Strategic Program

J. Greene et al arXiv:2206.14908

The primary objective of the PFS galaxy evolution survey is to bridge between large-scale structure and galaxy evolution.

​

- Physics cosmic reionization w/ LαE & 21cm

- Tomography of gas and DM

- Small scale test of the structure growth

- environmental dependent evolution

- and more...

Expect for CD3 to play a major role to contribute producing science!

LSST Scientific Motivations

Takeo Higuchi

Expert: Masahiro Takada

LSST: Legacy Survey of Space and Time;�was known as Large Synoptic Survey Telescope

• Studying dark energy and dark matter distributions by measuring weak gravitational lensing, baryon acoustic oscillations, and photometry of type Ia supernovae.
• Mapping small objects in the solar system.
• Detecting transient astronomical events (novae, supernovae, gamma-ray bursts, quasar variability, and gravitational lensing).
• Mapping the Milky Way.

LSST is a visible and IR telescope featured by the photographing power of the entire available sky in a few nights.

Apparatus for LSST

Takeo Higuchi

Expert: Masahiro Takada

• Visible and IR telescope.
• Wide survey area per photograph (9.6 deg²) with�a 8.4 m primary mirror.
• Mosaic of 189 CCD detectors each with 16M pixels.�→ produced data size = 15-20 TB/night, 60 PB in the LSST lifetime (10 years).

​

• Status: the mirror is ready; onsite camera commissioning is planned in 2024.

Data transfer speed between Chile↔︎US(NCSA) = 100 Gbps.

Possible collaboration with CD3

• Statistical analysis of the LSST’s huge data.�→ Physicists' involvement in the LSST project is encouraged (Takada-san).
• Deployment of machine learning technology for simulation studies.

Masaki Yamashita

Expert: Linda Blot

• Space telescope from ESA launched in July 2023
• 6 year survey of the whole extra-galactic sky
• Main goal: constraining dark energy and modified gravity beyond General Relativity
• Legacy science: galaxy evolution, Milky Way, transients, etc.

Cosmological probes

​

Weak lensing: statistical distortion of galaxy shapes due to the gravitational lensing caused by the intervening matter

Galaxy clustering: correlations of 3D position of galaxies

Masaki Yamashita

Expert: Linda Blot

Data challenges: Image level: 20 Petabytes, catalogue level: 1.5 billion galaxy shapes and 35 million galaxy spectra. ML is essential in some steps and can speed up many parts of the pipelines.

2 instruments:

​

VIS: imaging in the visible band

NISP: spectra in the near infrared

Toshiya Namikawa

Expert: Tad Takahashi

• The origins of Galactic positrons

• Uncover the sites of nucleosynthesis in the Galaxy: observing nuclear-lines from stellar and supernova nucleosynthesis (²⁶Al, ⁶⁰Fe, ⁴⁴Ti)

• Pioneering studies of gamma-ray polarization: Explore emission mechanisms of GRBs, black-hole systems, AGN

• Find counterparts to multi-messenger sources.

Science goals

The 511 keV gamma-ray line, signature of positron annihilation, has been discovered, but its origin from galactic center is unknown. COSI will produce the first direct image of the 511 keV line from the galactic center, as well as measure line profiles in these regions.

(INTEGRAL/SPI Galactic center map, Bouchert et al. 2010)

• Compton Spectrometer and Imager (COSI),
• A soft gamma-ray survey telescope (0.2-5 MeV)
• Launch in 2027 as a NASA small astrophysics mission

Nucleosynthesis products, individual sources, DM, …

Toshiya Namikawa

Expert: Tad Takahashi

• Improved sensitivity, spectral resolution, angular resolution, and sky coverage

• The COSI employs a novel Compton telescope design, utilizing sixteen 3-D imaging, high spectral resolution germanium detectors (GeDs).

An incoming gamma ray scatters and ends up in a photoabsorption event. By recording position and energy of each scattering, one can constrain the origin of the photon to an annulus on the sky. Observerving many photons from the same source, the circles overlap at a point towards the source.

Linear polarization can be also measured by its dependence on azimuthal scattering angle

(Kierans et al. 2022)

Instruments

Possible synergies with CD3

• Machine learning approaches will be an essential part of future gamma-ray missions: �Reconstruction of Compton events with ML
• IPMU is contributing leadership and direction to the Dark Matter working group �(T. Takahashi, S. Matsumoto, T. Melia)

The Lite (Light) satellite for the study of B-mode polarization and Inflation from cosmic background Radiation Detection

• Aims to find evidence of �cosmic inflation through �observations of the cosmic microwave background
• Can explain the uniformity of the CMB across the whole sky, which should not all be causally connected (horizon problem)
• Inflation is theorized to create unique primordial gravitational waves that imprint a distinct pattern in the CMB’s polarization signal called “primordial B-modes”, which LiteBIRD will try to detect

Patrick de Perio

Expert: Tomotake Matsumura

• Satellite with 3 mm-wave telescopes
• ~4.5k transition edge sensors across 15 observational frequency bands between 34 – 448 GHz
• Cooled to 5 K low temperatures to reduce thermal noise
• Capable of distinguishing the CMB and foreground
• Many of these frequencies are blocked by our atmosphere
• Deploy at the Sun-Earth L2 Lagrangian point in space
• Uniformly observe the entire sky for 3 years from ~FY2032

Prototype

@ IPMU 1F

• Polarization modulation units (PMU) using continuously rotating half-wave plates (HWP) modulate the polarized light and overcover the instrumental excess noise at low frequency where the signal also appears

• Rotatable cryostat (1.5 W cooler, 4K)
• LB only allows 40 mW!
• Windows on top and bottom to transmit mm-wave signal to test the optical device at the cryogenic temperature

• Leaderships in analysis: LiteBIRD x CMB-S4, Foreground Modelling, Component Separation
• CD3 prospects: developing new analyses and data pipelines

Science Goals:

• Probe the CMB polarization anisotropies:
• Large scales: primordial gravitational waves from inflation
• Small scales: CMB lensing, SZ, amplitude of matter perturbations σ₈(z), Neff, sum of neutrino masses, probe reionization… These SO observations promise to provide breakthrough discoveries in fundamental physics, cosmology and astrophysics.

Overview: 10 time better sensitivity and 5 time better angular resolution than the Planck satellite

• Atacama desert in Chile ~5200m, high and dry: better atmospherical conditions
• 2 types of telescopes: (SAT 10% of the sky, LAT 50 % of the sky)
• 3 SATs 0.5m lenses, rotating HWP, FOV:35deg, 0.5° resolution, >30,000 detectors (TES)
• LAT: 6m mirror, >30 000 TES, FOV:7deg, ~arcmin resolution (Operation at 0.1 K)
• 6 frequency bands 30–280 GHz (1–10mm)

Tad Takahashi

Expert: Baptiste Jost

15m

14m

LAT

SAT

B-mode polarization

Power spectrum

Tad Takahashi

Expert: Baptiste Jost

Construction is almost done! 2SATs being calibrated, last one on its way

3 more SATs are planed thanks to SO Japan, and SO:UK!

Data Analysis Challenges:

• Huge stream of time ordered data to reduce analyse
• Foreground cleaning
• Delensing
• Many systematic effects to model: atmosphere, ground signal, polarization angles, detector gain…

Masahito Yamazaki

Expert: Toshiya Namikawa

Science targets include:

​

• Primordial GW (tensor-to-scalar ratio r)
• Neff (light relic, such as axion, sterile neutrinos, dark radiation,...)

Need delicate analysis inside foreground-dominated regimes

Masahito Yamazaki

Expert: Toshiya Namikawa

• Located both in

Chile and South Pole

• ML is not completely obvious for CMB? However e.g. pipelines for foreground removal & systematics estimation

Large-area survey: Neff

small-area survey: r