Big Techday 23: Generative Data Calibrates the Sky [EN] - Dr Tamás Varga, LMU Munich
Generative Data Calibrates the Sky
Dark energy - the accelerating expansion of the Universe - is one of the key unsolved mysteries of physics today. A prominent way to understand cosmic expansion and growth is through vast surveys of the night sky where hundreds of millions of galaxies are photographed and catalogued, searching for subtle correlations in their positions and shapes. At LMU and the Max Planck for extraterrestrial Physics we develop powerful machine learning and Bayesian inference pipelines to extract insight from raw data. But calibrating and validating these tools cannot be done with real observations alone! Instead we build fair and statistically valid generative synthetic data models to characterise and forecast performance for yet to be seen astrophysical measurements. In my talk I will outline the key machine learning methods which power our work within the Dark Energy Survey (DES) and the Legacy Survey of Space and Time (LSST) and highlight their real world applications which could bring business value.
About the speaker
Dr. Tamás Varga: Dr. Tamás Norbert Varga is a research scientist at LMU and Max Planck Institute for extraterrestrial Physics where he is responsible for designing the generative synthetic data pipeline which will help power the largest cosmological survey ever constructed, the Legacy Survey of Space and Time (LSST). Cosmology provides a unique insight into a world where generative data tools have already been ubiquitous for years, and where the focus has therefore shifted from generation to ensuring the certification and statistical validity of AI methods. As a member of both LSST and the Dark Energy Survey (DES), Tamás worked at the frontier of data intensive cosmology research, in which he received his PhD in 2020 at LMU.
Видео Big Techday 23: Generative Data Calibrates the Sky [EN] - Dr Tamás Varga, LMU Munich канала TNG Technology Consulting GmbH
Dark energy - the accelerating expansion of the Universe - is one of the key unsolved mysteries of physics today. A prominent way to understand cosmic expansion and growth is through vast surveys of the night sky where hundreds of millions of galaxies are photographed and catalogued, searching for subtle correlations in their positions and shapes. At LMU and the Max Planck for extraterrestrial Physics we develop powerful machine learning and Bayesian inference pipelines to extract insight from raw data. But calibrating and validating these tools cannot be done with real observations alone! Instead we build fair and statistically valid generative synthetic data models to characterise and forecast performance for yet to be seen astrophysical measurements. In my talk I will outline the key machine learning methods which power our work within the Dark Energy Survey (DES) and the Legacy Survey of Space and Time (LSST) and highlight their real world applications which could bring business value.
About the speaker
Dr. Tamás Varga: Dr. Tamás Norbert Varga is a research scientist at LMU and Max Planck Institute for extraterrestrial Physics where he is responsible for designing the generative synthetic data pipeline which will help power the largest cosmological survey ever constructed, the Legacy Survey of Space and Time (LSST). Cosmology provides a unique insight into a world where generative data tools have already been ubiquitous for years, and where the focus has therefore shifted from generation to ensuring the certification and statistical validity of AI methods. As a member of both LSST and the Dark Energy Survey (DES), Tamás worked at the frontier of data intensive cosmology research, in which he received his PhD in 2020 at LMU.
Видео Big Techday 23: Generative Data Calibrates the Sky [EN] - Dr Tamás Varga, LMU Munich канала TNG Technology Consulting GmbH
Показать
Комментарии отсутствуют
Информация о видео
2 августа 2023 г. 18:05:36
00:51:13
Другие видео канала
![Big Techday 23: Glean: Query your Code [EN] - Simon Marlow, Meta Research](https://i.ytimg.com/vi/eFglEyb-KfM/default.jpg)
![Big Techday 23: Cognitive Robots [EN] - Dr. Alexander Blass, NEURA Robotics](https://i.ytimg.com/vi/in9tvbOQvQU/default.jpg)
![Big Techday 23: Taming the Home Connect Microservice Jungle [EN] - T. Richer, BSH & T. Masur, TNG](https://i.ytimg.com/vi/zRZaUzQdLGc/default.jpg)
![Big Techday 23: sAI Cheese! - Professionelle Profilbilder mit Generativer KI [DE] - Endres & Mayer](https://i.ytimg.com/vi/mqovAkY2_MA/default.jpg)
![Big Techday 23: Der Roboter in meinem Wohnzimmer [DE] - Matthias Lau, Heureka Labs](https://i.ytimg.com/vi/HVq2pam2nbE/default.jpg)
![Big Techday 23: Domain Sniping – Hunting for the World's Shortest Domains [EN] - E. Goslawski, TNG](https://i.ytimg.com/vi/46dVyhcdul4/default.jpg)
![Big Techday 23 & MuniHac 2023: Haskell at the Heart of Terabit Laser Communication [EN] - C. Baaij](https://i.ytimg.com/vi/sruXs0GydE0/default.jpg)
![Big Techday 23: Veränderung des Schachs durch Technik [DE] - Vincent Keymer, Schach-Großmeister](https://i.ytimg.com/vi/IOQNVN4aX7s/default.jpg)
![Big Techday 23: Digitalization in Healthcare [DE] - Prof. Gotthardt & Dr. Pech](https://i.ytimg.com/vi/FfXLIBt3th8/default.jpg)
![Big Techday 23: Erkennung von Herzerkrankungen mit KI [DE] - Thomas Fechner, Cardisio GmbH](https://i.ytimg.com/vi/_15OgaEpufk/default.jpg)
![Big Techday 23: Trunk Based Development in der Praxis [DE] - C. Deger, RIO & M. Spanier, TNG](https://i.ytimg.com/vi/ExhdAhLrWaI/default.jpg)
![Big Techday 22: My Pursuit of the Fastest Drive in History [EN] - A.Toman, Cannonball Run Winner](https://i.ytimg.com/vi/Acf8A_N0gZ8/default.jpg)
![Big Techday 22: Usable Security Research @ KIT (DE) - Prof. Dr. Melanie Volkamer, KIT](https://i.ytimg.com/vi/us79q2ACTvY/default.jpg)
![Big Techday 22: Lightspeed to SaaS ... From a Delphi Monolith to the Cloud [EN] - S. Alber, Zvoove](https://i.ytimg.com/vi/A6sFY8rmcqM/default.jpg)
![Big Techday 22: Next Level Schafkopf-KI [DE] - Franz Pfeifroth, Alexander Sedlmayr](https://i.ytimg.com/vi/CeTgUeWEvWQ/default.jpg)
![MuniHac 2022 - Sunday, October 9 - Nicolas Wu: Modular Programming with Effects](https://i.ytimg.com/vi/wk3Qj15WPpk/default.jpg)
![MuniHac 2022 - Saturday, October 8 - Michael Sperber: How to Deep Learn with Categories](https://i.ytimg.com/vi/y-McNDBbS7M/default.jpg)
![MuniHac 2022 - Saturday, October 8 - Ben Gamari: Improvements in observability in GHC 9.6](https://i.ytimg.com/vi/vCv1vnY87NI/default.jpg)
![MuniHac 2022 - Saturday, October 8 - Krzysztof Gogolewski: Quines and metaprogramming](https://i.ytimg.com/vi/Rnji4rZT51s/default.jpg)
![MuniHac 2022 - Cass Alexandru: Structured Traversals for (Mutually) Recursive Algebraic Data Types](https://i.ytimg.com/vi/Xk_inDd0Pgk/default.jpg)