CPLUOS Astrophysics team meeting

Friday 30 Sept 2022, 14:00 → 15:20 Asia/Seoul

물리학과

https://uos-ac-kr.zoom.us/j/8264902652

COSMOLOGY

1 Hyeonmo's Report - Introduction of GALIC code

Speaker: Hyeonmo Koo (University of Seoul)

20220930.pptx

Introduction of GALIC code

• GALIC head-on collision
  • Colorbar minimum = 5
  • GALIC cord has 10 conditions.
  • Color = density
• Try Dice code.

2 Young's report. FoF and MST algorithm

Speaker: Young Ju

FoF and MST Algorithm

• Mulguisin clustering algorithm
• Polar angles are different.

3 Seyeon's Report - CLML

Speaker: Se Yeon Hwang (Universe of Seoul)

Cosmology with Large scale structure using Machine Learning (CLML)

• No progress

4 Sumi's report

Speaker: Sumi Kim (University of Seoul)

Cosmology of High-Order Statistics (CHOA)

• Redshift split → The advice: Using three redshift bins
• MLPRegressor n = 512
• Layers = (n, n, 2n, 2n, 2n, 2n, n, n)
• This week
  • 3 pcf is currently running now (1/3 finished).
  • No making theoretical part slides for KAS presentation

5 John Suarez's report

Join Photo+Spect Data

• Photometric normalized.
  • Epoch: 10
  • Learning rate: 0.01
  • All redshift (including Z > 0.5)
    • (Model, Learning rate, Batch size, Loss (MSE))
      • (resnet50, 0.01, 32, 0.614)
• Photometric data models
  • Redshifts z < 0.5
  • Resnet +3 linear [512, 256, 1] → Redshift
    • (Model, Learning, Batch size, Epochs, Loss)
      • (resnet50, 0.01, 32, 300, 0.828)
• To do
  • Spectroscopic data models
  • Photometric data models
  • Data set
    • In progress a data set from GUADALUPE (Last DESI DR)

6 Hannah's Report

Speaker: Hannah Jhee (University of Seoul)

Halos infalling a cluster through filaments

• Snapshot number
  • Cosine graph
    • Nearest point
    • Most dense
    • Perpendicular method → Red line is mistake. → cos0 ≠ 0
    • The process: Halo → Nearest filament → Nearest+densest (?) filament

7 Dr.Sabiu's Report

Speaker: Dr Cristiano Sabiu (University of Seoul)

Deep Learning the Deep Sky: Recovering low surface brightness objects with machine learning

• No progress

8 Discussion