科学研究
-
2023.06.29
Exploring the Universe with Space Transition-edge Sensor: from Cosmic Microwave to X-ray Background
One commonly accepted idea for the origin of the large-scale structure in the Universe is that the minute quantum fluctuations were stretched to macroscopic scales as a consequence of nearly exponential expansion of spacetime during the “cosmic inflation”. Cosmic inflation predicts primordial density fluctuations that are consistent with the observed CMB temperature fluctuations. It also pred...
详情 -
2023.06.28
Kamioka CryoLab: A Future Low-background Underground Facility for Cryogenic Light Dark Matter Searches
Dark matter — established via various cosmological and astronomical observations — is a significant constituent of our Universe and remains one of the most outstanding mysteries of modern physics. The mass range of potential dark matter candidates covers more than 30 orders of magnitude. In the past, researchers have primarily focused on searching for GeV-TeV dark matter (WIMP) via nuclear re...
详情 -
2023.06.28
Breaking bad degeneracies with Love relations: Improving gravitational-wave measurements through universal relations
The distance-inclination degeneracy limits gravitational-wave parameter estimation of compact binary mergers. Such a degeneracy can be partially broken by including higher-order modes or precession when modeling the waveform of a binary that contains a black hole. But what about binary neutron stars, for which these effects are suppressed? In this talk, I will introduce a new parameterization o...
详情 -
2023.06.19
Deep Learning the Local Group
As the witnesses of the early Universe, dwarf galaxies, especially the low-mass ones (total mass ≤ 10^9 M⊙), carry the imprint of essential physical processes, from the formation of the first stars to the formation of the first galaxies, which shaped the universe into the one we observe today. I will show the power of searching dwarf galaxy debris in the low-metallicity Milky Way, and how the...
详情 -
2023.06.16
Flow Annealed Kalman Filters: Gradient-free inference for expensive forward models
For many scientific applications the forward model consists of an expensive black-box simulator without easy access to gradients of the forward model. In such a situation, inference can be challenging, with standard gradient-free sampling algorithms requiring an intractably large number of serial model evaluations. In this talk I will introduce the flow annealed Kalman filter for approximate Ba...
详情