Esperanza Lopez Aguilera - Using a Bayesian Neural Network in the Detection of Exoplanets
Download as PPTX, PDF1 like667 views
This document discusses using a Bayesian neural network to classify light curves from the Transiting Exoplanet Survey Satellite (TESS) mission to identify exoplanet candidates. It describes challenges in classifying large numbers of light curves, and how a Bayesian neural network approach provides probabilistic predictions and confidence levels to help identify promising exoplanet candidates while avoiding many false positives seen in other methods. The Bayesian network achieved 91% accuracy and 83% precision in tests on simulated TESS data.
![Frossie Economou & Angelo Fausti [Vera C. Rubin Observatory] | How InfluxDB H...](https://cdn.slidesharecdn.com/ss_thumbnails/veracrubininfluxdays2020-201111202349-thumbnail.jpg?width=560&fit=bounds)
Ad
More Related Content
Similar to Esperanza Lopez Aguilera - Using a Bayesian Neural Network in the Detection of Exoplanets (20)
More from MLconf (20)
Ad
Recently uploaded (20)
Ad
Esperanza Lopez Aguilera - Using a Bayesian Neural Network in the Detection of Exoplanets
- 1. 1The World’s Fastest Time-Series Database Esperanza López Aguilera 29 March 2019 Using a Bayesian Neural Network in the Detection of Exoplanets
- 2. 2The World’s Fastest Time-Series Database ● The World’s Fastest Time-Series Database ○ In-memory computing ○ Streaming analytics ● Q Language ○ Functional ○ Array based ○ Primitive temporal datatypes ○ Tables are the first class datatype ○ Qsql ○ Lambda architecture out of the box ○ Scarily efficient ○ Syntax highlighting for q What’s kdb+?
- 3. 3The World’s Fastest Time-Series Database ● The World’s Fastest Time-Series Database ○ In-memory computing ○ Streaming analytics ● Q Language ○ Functional ○ Array based ○ Primitive temporal datatypes ○ Tables are the first class datatype ○ Qsql ○ Lambda architecture out of the box ○ Scarily efficient ○ Syntax highlighting for q What’s kdb+? Extremely fast Elegant and concise
- 4. 4The World’s Fastest Time-Series Database Applications What do they have in common? Data - Time series Machine learning solutions
- 5. 5The World’s Fastest Time-Series Database NASA Frontier Development Lab ● Applied AI research accelerator ● Hosted by: ○ SETI Institute ○ NASA Ames Research Center ● 2-months programme ● 7 challenges ● Involved in 2 challenges: ○ Space weather: ■ How solar activity impacts Earth ■ Paper: https://code.kx.com/q/wp/space-weather/ ○ Exoplanets: ■ Find new planet candidates ■ Paper: https://code.kx.com/q/wp/exoplanets/
- 6. 6The World’s Fastest Time-Series Database Exoplanets Challenge - TESS ● Launched in April 2018 ● 2 years mission ● 26 sectors: ○ 27 days per sector ● Objective: Discovering new exoplanets in orbit around the brightest stars in the solar neighborhood
- 7. 7The World’s Fastest Time-Series Database Exoplanets challenge - Transits How do we detect exoplanets? Transits
- 8. 8The World’s Fastest Time-Series Database But it’s not so easy ... Background Eclipsing Binaries Eclipsing Binaries Stellar activity
- 9. 9The World’s Fastest Time-Series Database Data Images taken at a given frequency Target stars Optimal set of pixel representing each star Aggregate brightness extracted Remove noise, trends and other factors ● Simulated data ○ 4 sectors ● 64, 000 target stars ● Strong signal found in 9,139 stars ● 19,577 TCEs or planet candidates ● Optimal parameters inferred: ○ Epoch ○ Period ○ Duration ○ ... ● Issue: Many false positives Threshold Crossing Events Corrected flux - Light curve
- 10. 10The World’s Fastest Time-Series Database Light curves Where is the planet?
- 11. 11The World’s Fastest Time-Series Database Light curves What? Why?
- 12. 12The World’s Fastest Time-Series Database Light curves This is confusing
- 13. 13The World’s Fastest Time-Series Database Light curves - Local view Now, I know
- 14. 14The World’s Fastest Time-Series Database Classification techniques ● Humans looking at light curves ○ Statistical methods used ○ Too many hours ● Complex models ○ Several inputs ○ Time-consuming ○ Intensive preprocessing
- 15. 15The World’s Fastest Time-Series Database Benchmark model ● 77% accurate ● Very low precision ○ 47% ● Uncertainty? ● Confidence? Linear classifier
- 16. 16The World’s Fastest Time-Series Database Bayesian Neural Network Stochastic model + Neural Network Probabilistic confidence on predictions ● Weights follow a distribution ● Train parameters instead of weights ● Result: Distribution of probabilities ● Several criteria for decision making ○ Standard deviation ○ Mean ○ ...
- 17. 17The World’s Fastest Time-Series Database Bayesian Neural Network Oversampling - Random sample of the positive class Build Network - Define architecture and parameters EmbedPy
- 18. 18The World’s Fastest Time-Series Database Results - Performance ● Outputs of the BNN: ○ Probability of being a planet ○ Sample of size 500 per input ● Decision based on: ○ Average probability ○ P > 0.5 ⇒Planet ○ Flexibility ● Metrics: ○ Accuracy: 91% ○ Precision: 83% ○ Sensitivity: 68% Probability sample of one TCE
- 19. 19The World’s Fastest Time-Series Database Results - Confidence
- 20. 20The World’s Fastest Time-Series Database Results - Examples
- 21. 21The World’s Fastest Time-Series Database Esperanza López Aguilera Machine Learning Engineer, Kx [email protected] [email protected] Useful links Machine learning: https://code.kx.com/q/ml/ Kx Github: https://github.com/KxSystems/ Join the conversation: twitter.com/kxsystems linkedin.com/company/kx-systems/ facebook.com/kxsystems instagram.com/kxsystems Thanks Contact details