Characterizing Ext xtragalactic Pre-Main- Sequence Stars wit ith - - PowerPoint PPT Presentation

▶

Nov 15, 2023 152 likes •334 views

Characterizing Ext xtragalactic Pre-Main- Sequence Stars wit ith Machine and Deep Learnin ing Techniques Victor Francisco Ksoll Understanding the Nearby Star-forming Universe with JWST 27.08.2019 Star Forming Regions in the Large Magellanic

SLIDE 1

Characterizing Ext xtragalactic Pre-Main- Sequence Stars wit ith Machine and Deep Learnin ing Techniques

Understanding the Nearby Star-forming Universe with JWST 27.08.2019

Victor Francisco Ksoll

SLIDE 2

Star Forming Regions in the Large Magellanic Cloud

Data: Hubble Tarantula Treasury Project HTTP Region: Tarantula Nebula Sources: > 800.000 Observed Area: 210 pc x 180 pc Filters: F275W, F336W, F555W, F658N, F775W, F110W, F160W Data: Measuring Young Stars in Space and Time MYSST Region: Star forming Complex Nebula N44 Sources: > 400.000 Observed Area: 135 pc x 190 pc Filters: F555W, F814W

Credit: NASA, ESA, E. Sabbi (STScI) Credit: D. Gouliermis and MYSST

SLIDE 3

Supervised ML classification of f stars

CMD Spatial Map

[mag] [mag]

SLIDE 4

Supervised ML classification of f stars

CMD Spatial Map

[mag] [mag]

Training Set: R136

SLIDE 5

Supervised ML classification of f stars

CMD Spatial Map

[mag] [mag]

SLIDE 6

Supervised ML classification of f stars

ML-based classification of PMS stars with RFs and SVM techniques

Ksoll, et al 2018, MNRAS, Volume 479, Issue 2, p2389-2414

CMD Spatial Map

[mag] [mag]

SLIDE 7

Unsupervised Clustering of f stars

Contour Density Based Clustering Analysis

Spatial Density Map

Ksoll, Gouliermis et al., in prep

Dendrogram

SLIDE 8

N44 Classification

CMD

Spatial Map

SLIDE 9

Goals and Methods

Identifying PMS-Stars:

Aim: Separate Young from Old Stars
Task: Supervised Classification
Training set: Observations
Photometry of known Clusters
Algorithms:

Support Vector Machine (SVM), Random Forest (RF)

Characterizing PMS-Stars

Aim: Determine Physical Parameters of Stars
Task: Regression
Training set: Theoretical Models
Isochrones from Stellar Evolution Models
Algorithms:

Invertible Neural Network (INN)

SLIDE 10

Physical parameter regression with an In Invertible Neural Network (I (INN)

SLIDE 11

Physical parameter regression with an In Invertible Neural Network (I (INN)

SLIDE 12

Physical parameter regression with an In Invertible Neural Network (I (INN)

Ardizzone, et al 2018, arXiv:1808.04730

SLIDE 13

Physical parameter regression with an In Invertible Neural Network (I (INN)

Current Training Progress

[yr] [Msun] [yr] [Msun]

SLIDE 14

IN INN Application Test Case Westerlund 2

Credit: NASA, ESA, Antonella Nota (ESA/STScI), E. Sabbi (STScI)

Photometry in F814W and F160W Cluster Sources: ca. 6200 Gas extinction map: Zeidler et al. 2015, AJ. Age (Zeidler et al. 2015): 0.5 – 2.0 Myr 1.5 Myr Isochrone

SLIDE 15

IN INN Application Test Case Westerlund 2

Credit: NASA, ESA, Antonella Nota (ESA/STScI), E. Sabbi (STScI)

Photometry in F814W and F160W Cluster Sources: ca. 6200 Gas extinction map: Zeidler et al. 2015, AJ. Age (Zeidler et al. 2015): 0.5 – 2.0 Myr

SLIDE 16

IN INN Application Test Case Westerlund 2

Credit: NASA, ESA, Antonella Nota (ESA/STScI), E. Sabbi (STScI)

Photometry in F814W and F160W Cluster Sources: ca. 6200 Gas extinction map: Zeidler et al. 2015, AJ. Age (Zeidler et al. 2015): 0.5 – 2.0 Myr

SLIDE 17

IN INN Application Test Case Westerlund 2

Credit: NASA, ESA, Antonella Nota (ESA/STScI), E. Sabbi (STScI)

SLIDE 18

Thank you!

The entire Field of View of the MYSST survey

INN Framework: github.com/VLL-HD/FrEIA