Scaling-up Data Science
In the “good ol’ days” – if they have ever existed – we used to have models aimed at explaining phenomena via a limited number of parameters and we could test them using a small amount of data. When we collected new data, we only had to feed them into the model and compute the outcome.
Nowadays, statistical and machine learning models can have millions of parameters and we can collect billions of heterogeneous datapoints coming from different sources, and no computer in the world is able to process such quantities in a reasonable amount of time. That’s what computational algorithms are for: they are processes that come to around the same results of the original model, but in a simpler and faster way.
There are some issues, though. We don’t always exactly understand why a computational algorithm works and, if it does, we can’t be sure it will work as well with different or considerably larger datasets.
“This lack of understanding results in the routine use of inefficient and largely suboptimal algorithms, and makes the design of efficient algorithms for practically used models something of an art,” said Giacomo Zanella, Assistant Professor at Bocconi Department of Decision Sciences.
Zanella obtained a €1.5 ERC Starting Grant from the European Research Council (ERC) to better understand computational algorithms for large-scale probabilistic models, thus making their design not an art, but a science. The project (PrSc-HDBayLe - Provable scalability for high-dimensional Bayesian Learning) aims to single out the most promising algorithms using rigorous and innovative mathematical techniques, and to produce guidelines to improve them and develop new ones.
Subscribe to our channel ➤ @UniBocconi
Visit our website: https://www.unibocconi.eu/
Follow our social media accounts:
LinkedIn ➤ https://linkedin.com/school/universita-bocconi/
Facebook ➤ https://facebook.com/unibocconi
Twitter ➤ https://twitter.com/Unibocconi
Instagram ➤ https://instagram.com/unibocconi/
#Bocconi #BocconiUniversity #UniBocconi
Видео Scaling-up Data Science канала Bocconi University
Nowadays, statistical and machine learning models can have millions of parameters and we can collect billions of heterogeneous datapoints coming from different sources, and no computer in the world is able to process such quantities in a reasonable amount of time. That’s what computational algorithms are for: they are processes that come to around the same results of the original model, but in a simpler and faster way.
There are some issues, though. We don’t always exactly understand why a computational algorithm works and, if it does, we can’t be sure it will work as well with different or considerably larger datasets.
“This lack of understanding results in the routine use of inefficient and largely suboptimal algorithms, and makes the design of efficient algorithms for practically used models something of an art,” said Giacomo Zanella, Assistant Professor at Bocconi Department of Decision Sciences.
Zanella obtained a €1.5 ERC Starting Grant from the European Research Council (ERC) to better understand computational algorithms for large-scale probabilistic models, thus making their design not an art, but a science. The project (PrSc-HDBayLe - Provable scalability for high-dimensional Bayesian Learning) aims to single out the most promising algorithms using rigorous and innovative mathematical techniques, and to produce guidelines to improve them and develop new ones.
Subscribe to our channel ➤ @UniBocconi
Visit our website: https://www.unibocconi.eu/
Follow our social media accounts:
LinkedIn ➤ https://linkedin.com/school/universita-bocconi/
Facebook ➤ https://facebook.com/unibocconi
Twitter ➤ https://twitter.com/Unibocconi
Instagram ➤ https://instagram.com/unibocconi/
#Bocconi #BocconiUniversity #UniBocconi
Видео Scaling-up Data Science канала Bocconi University
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
![Bocconi meets Munich 30'' #bocconi #unibocconi #bocconiuniversity](https://i.ytimg.com/vi/LhfQJJPmKac/default.jpg)
![Bocconi meets Munich 15'' #bocconi #bocconiuniversity #unibocconi](https://i.ytimg.com/vi/N_OzowdU4zA/default.jpg)
![Bocconi meets Paris 30'' #bocconi #unibocconi #bocconiuniversity](https://i.ytimg.com/vi/mE5DeeQbCd4/default.jpg)
![Bocconi meets Paris 15'' #bocconi #bocconiuniversity #unibocconi](https://i.ytimg.com/vi/4z70MGJXN3k/default.jpg)
![What Cognitive Economics Is About](https://i.ytimg.com/vi/vm-HJBesFWQ/default.jpg)
![Take the cap!](https://i.ytimg.com/vi/ivREropMfrI/default.jpg)
![Highlights from the Bocconi Summer School 2022](https://i.ytimg.com/vi/kD7YM2NlwfA/default.jpg)
![Discover the Bocconi Summer School Experience](https://i.ytimg.com/vi/_eRPkiYeT4I/default.jpg)
![You Can Watch This Video Only Once](https://i.ytimg.com/vi/KGBc8q7_yIc/default.jpg)
![Best Wishes to all of you!](https://i.ytimg.com/vi/Sb61mf9rEt4/default.jpg)
![Bocconi. Italy. Europe. A tribute to Mario Monti. Highlights from the event](https://i.ytimg.com/vi/vHYKkMfGTc8/default.jpg)
![Bocconi, Italy, Europe. A tribute to Mario Monti](https://i.ytimg.com/vi/2vlGtxDK8bo/default.jpg)
![Financial Markets Beyond Stock Exchanges](https://i.ytimg.com/vi/pjt6r0TzMeU/default.jpg)
![Promoting Research That Fits the Needs of Our Society](https://i.ytimg.com/vi/0F6nuGJXQ_E/default.jpg)
![Bocconi BA in Global Law](https://i.ytimg.com/vi/AFDfuE95UPM/default.jpg)
![Executive Chat with Elena Goitini, CEO of BNL and Head of BNP Paribas Group in Italy](https://i.ytimg.com/vi/Pz0UjvEg2rs/default.jpg)
![2022-23 Academic Year Opening Ceremony | Inspired by Complexity](https://i.ytimg.com/vi/A-PTd7osxqA/default.jpg)
![Executive Chat with Davide Serra, Founder and CEO of Algebris Investments](https://i.ytimg.com/vi/HmVP-qhYL-E/default.jpg)
![1800sec – Dare to ask. With Maurizio Dallocchio and Paolo Scaroni](https://i.ytimg.com/vi/L4H6yo6Jvqg/default.jpg)
![Short-term Debt Is More Costly Than You Think](https://i.ytimg.com/vi/XHkBuSVn0ig/default.jpg)