Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
Rewinder
Search
Adrian Price-Whelan
October 30, 2014
Science
0
87
Rewinder
Presented at the Gaia Challenge 2 at the MPIA in Heidelberg.
Adrian Price-Whelan
October 30, 2014
Tweet
Share
More Decks by Adrian Price-Whelan
See All by Adrian Price-Whelan
the Astropy project - Flatware
adrn
1
220
the dynamic Milky Way in the Gaia era
adrn
1
200
The Astropy Project
adrn
1
120
Git and version control
adrn
1
110
Chaos and stellar streams
adrn
1
170
Software testing
adrn
0
200
Local Group Astrostatistics
adrn
1
86
100% Outer Space
adrn
1
150
SMHASH telecon 03/2015
adrn
0
100
Other Decks in Science
See All in Science
化学におけるAI・シミュレーション活用のトレンドと 汎用原子レベルシミュレーター: Matlantisを使った素材開発
matlantis
0
480
Improving Search @scale with efficient query experimentation @BerlinBuzzwords 2024
searchhub
0
280
3次元点群を利用した植物の葉の自動セグメンテーションについて
kentaitakura
2
930
統計的因果探索: 背景知識とデータにより因果仮説を探索する
sshimizu2006
3
690
02_西村訓弘_プログラムディレクター_人口減少を機にひらく未来社会.pdf
sip3ristex
0
190
The Incredible Machine: Developer Productivity and the Impact of AI
tomzimmermann
0
570
局所保存性・相似変換対称性を満たす機械学習モデルによる数値流体力学
yellowshippo
1
180
位相的データ解析とその応用例
brainpadpr
1
1k
はじめてのバックドア基準:あるいは、重回帰分析の偏回帰係数を因果効果の推定値として解釈してよいのか問題
takehikoihayashi
2
1.4k
20分で分かる Human-in-the-Loop 機械学習におけるアノテーションとヒューマンコンピューターインタラクションの真髄
hurutoriya
5
2.8k
理論計算機科学における 数学の応用: 擬似ランダムネス
nobushimi
1
410
Transformers are Universal in Context Learners
gpeyre
0
730
Featured
See All Featured
A better future with KSS
kneath
238
17k
The Cost Of JavaScript in 2023
addyosmani
47
7.5k
Building a Scalable Design System with Sketch
lauravandoore
462
33k
Visualization
eitanlees
146
15k
Testing 201, or: Great Expectations
jmmastey
42
7.2k
It's Worth the Effort
3n
184
28k
Thoughts on Productivity
jonyablonski
69
4.5k
The Cult of Friendly URLs
andyhume
78
6.2k
Rebuilding a faster, lazier Slack
samanthasiow
80
8.9k
The Power of CSS Pseudo Elements
geoffreycrofte
75
5.5k
Stop Working from a Prison Cell
hatefulcrawdad
268
20k
Save Time (by Creating Custom Rails Generators)
garrettdimon
PRO
29
1.1k
Transcript
REWINDER Adrian Price-Whelan (Columbia University) Kathryn Johnston, David Hogg, Sarah
Pearson, Andreas Küpper, Allyson Sheffield
Internal dynamics / disruption Underlying potential Streams encode info. about:
How to “fit” stream data? Need a generative model!
GRAVITY Progenitor Lagrange points
Stars Progenitor
Present Past Rewinder
None
Stripped star Progenitor
(Star orbit) - (Progenitor orbit)
x1 = ri · ˆ r prog ( t )
L prog (t), r prog (t), x3 = ri · ˆ L prog ( t ) x2 = ri · ˆ x2
Instantaneous orbital plane
Normalized by tidal radius
None
Incorrect potential
✓ The disruption is simple?
✓ At the (unknown) unbinding time for each star…
“The Pal 5 Challenge” with 4 stars: 3 leading arm
1 trailing arm
None
Previously… Price-Whelan et al. (2014) Nparams / 6Nstars Bug: Computationally
can only use a few stars Feature: We can infer the potential with a few stars!
Price-Whelan et al. (2014)
Now Use importance sampling to marginalize over “true” phase-space positions
of the stars Only params. of marginal likelihood are potential params., progenitor position, nuisance params. Price-Whelan et al. (in prep.)
Price-Whelan et al. (in prep.)
SMHASH Spitzer Merger History And Shape of the Halo RR
Lyrae stars: intrinsic 2% scatter in P-L relation in IR Johnston, Madore, Majewski, Besla, Bono, Clementini, Cioni, Cohen, Freedman, Grillmair, Kallivayalil, Kollmeier, Law, Marengo, Mondon, Nidever, Pietrzynski, Persson, Price-Whelan, Rich, Scowcroft, Seibert, Smith, Sesar, Soszynski, Thompson, Udalski, van der Marel (Madore & Freedman 2012) (+ Spitzer Cycle 9 work)
SMHASH Debris targets: ~30 stars in Orphan stream ~100 stars
in Sagittarius (Sgr) stream 2 Sgr globular clusters 20 Sgr core fields
None