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
False Start Detection in Elite Athletics
Search
Kevin Brosnan
October 18, 2016
Research
0
130
False Start Detection in Elite Athletics
Short Informal Mathematics (SIM) Talk given within the department at the University of Limerick
Kevin Brosnan
October 18, 2016
Tweet
Share
More Decks by Kevin Brosnan
See All by Kevin Brosnan
Automated Gating for Flow Cytometry
significantstats
0
200
False Starts in Athletics: Are they truly fair?
significantstats
0
100
False Starts in Athletics: Are they truly fair?
significantstats
0
90
A Markov Random Fields Approach to the Gating of Flow Cytometry Data
significantstats
0
140
A Markov Random Fields Approach to the Gating of Flow Cytometry Data
significantstats
0
140
Challenges for tertiary level mathematics tutors
significantstats
0
90
Elite Athletics: Is the false start disqualification rule appropriate?
significantstats
0
130
Quantile Regression
significantstats
0
150
Forward Modelling of UK Gas Prices
significantstats
0
58
Other Decks in Research
See All in Research
能動適応的実験計画
masakat0
2
710
Galileo: Learning Global & Local Features of Many Remote Sensing Modalities
satai
3
130
診断前の病歴テキストを対象としたLLMによるエンティティリンキング精度検証
hagino3000
1
120
2025年度人工知能学会全国大会チュートリアル講演「深層基盤モデルの数理」
taiji_suzuki
24
17k
SSII2025 [TS3] 医工連携における画像情報学研究
ssii
PRO
2
1.2k
一人称視点映像解析の最先端(MIRU2025 チュートリアル)
takumayagi
5
2.8k
心理言語学の視点から再考する言語モデルの学習過程
chemical_tree
2
510
集合間Bregmanダイバージェンスと置換不変NNによるその学習
wasyro
0
120
最適化と機械学習による問題解決
mickey_kubo
0
150
数理最適化と機械学習の融合
mickey_kubo
15
9k
生成的推薦の人気バイアスの分析:暗記の観点から / JSAI2025
upura
0
230
Delta Airlines® Customer Care in the U.S.: How to Reach Them Now
bookingcomcustomersupportusa
PRO
0
100
Featured
See All Featured
Embracing the Ebb and Flow
colly
86
4.8k
Visualization
eitanlees
146
16k
Documentation Writing (for coders)
carmenintech
73
4.9k
Reflections from 52 weeks, 52 projects
jeffersonlam
351
21k
How GitHub (no longer) Works
holman
314
140k
Site-Speed That Sticks
csswizardry
10
730
The Cult of Friendly URLs
andyhume
79
6.5k
Visualizing Your Data: Incorporating Mongo into Loggly Infrastructure
mongodb
47
9.6k
How to Ace a Technical Interview
jacobian
278
23k
Balancing Empowerment & Direction
lara
1
520
Responsive Adventures: Dirty Tricks From The Dark Corners of Front-End
smashingmag
251
21k
Designing Dashboards & Data Visualisations in Web Apps
destraynor
231
53k
Transcript
Fa l s e - St a r t D
e t e c t i o n i n E l i t e At h l e t i c s K e v i n B r o s n a n , U n i v e r s i t y O f L i m e r i c k S I M T a l k , 1 8 t h O c t o b e r 2 0 1 6
O u t l i n e … • Background
• Rule Changes • Exploratory Analysis • Modelling • Results • Outputs
B a ck g r o u n d …
R u l e C h a n ge s
… 1998 2004 2010 2016 Individual Warning
R u l e C h a n ge s
… 1998 2004 2010 2016 Individual Warning
R u l e C h a n ge s
… 1998 2004 2010 2016 Individual Warning
R u l e C h a n ge s
… 1998 2004 2010 2016 Individual Warning
R u l e C h a n ge s
… 1998 2004 2010 2016 Individual Warning Group Warning
R u l e C h a n ge s
… 1998 2004 2010 2016 Individual Warning Group Warning
R u l e C h a n ge s
… 1998 2004 2010 2016 Individual Warning Group Warning
R u l e C h a n ge s
… 1998 2004 2010 2016 Individual Warning Automatic Disqualification Group Warning
R u l e C h a n ge s
… 1998 2004 2010 2016 Individual Warning Automatic Disqualification Group Warning
E x p l o r a t o r
y … 2,310 1,007 1,303
E x p l o r a t o r
y … 2,310 1,007 1,303
M o d e l l i n g …
f ( RT|µ, , ⌧ ) = 1 ⌧ exp nµ ⌧ + 2 2⌧2 RT ⌧ o ✓RT µ 2 ⌧ ◆
M o d e l l i n g …
f ( RT|µ, , ⌧ ) = 1 ⌧ exp nµ ⌧ + 2 2⌧2 RT ⌧ o ✓RT µ 2 ⌧ ◆ µ = 400, = 100, ⌧ = 0
M o d e l l i n g …
f ( RT|µ, , ⌧ ) = 1 ⌧ exp nµ ⌧ + 2 2⌧2 RT ⌧ o ✓RT µ 2 ⌧ ◆ µ = 400, = 100, ⌧ = 0 µ = 400, = 100, ⌧ = 0 µ = 0, = 0, ⌧ = 150
M o d e l l i n g …
f ( RT|µ, , ⌧ ) = 1 ⌧ exp nµ ⌧ + 2 2⌧2 RT ⌧ o ✓RT µ 2 ⌧ ◆ µ = 400, = 100, ⌧ = 0 µ = 400, = 100, ⌧ = 0 µ = 0, = 0, ⌧ = 150 µ = 400, = 100, ⌧ = 0 µ = 0, = 0, ⌧ = 150 µ = 400, = 100, ⌧ = 50
M o d e l l i n g …
f ( RT|µ, , ⌧ ) = 1 ⌧ exp nµ ⌧ + 2 2⌧2 RT ⌧ o ✓RT µ 2 ⌧ ◆ µ = 400, = 100, ⌧ = 0 µ = 400, = 100, ⌧ = 0 µ = 0, = 0, ⌧ = 150 µ = 400, = 100, ⌧ = 0 µ = 0, = 0, ⌧ = 150 µ = 400, = 100, ⌧ = 50 µ = 400, = 100, ⌧ = 0 µ = 0, = 0, ⌧ = 150 µ = 400, = 100, ⌧ = 50 µ = 400, = 100, ⌧ = 150
R e s u l t s …
R e s u l t s …
R e s u l t s …
R e s u l t s …
O u t p u t s …
Thanks for Listening! Any Questions?