ÂÅÐÑÈß ÄËß ÑËÀÁÎÂÈÄßÙÈÕ
Enter
Login:
Password:
Forgot your password?
scientific activity
structureeducational projectsperiodicalsstaffpress centercontacts
ðóññêèé | english

Structural Models and Deep Learning

Joint IITP RAS, NRU HSE &Skoltech seminar  

November 1 (Tuesday), 1830, IITP, room 615   
 
Alexey Zaytsev (Skoltech, IITP) 
 
Minimax approach to modelling variable fidelity data (joint work with Engeny Burnaev) 
 
Engineering problems often involve data sources of variable fidelity with different costs of obtaining an observation. In particular, one can use both a cheap low fidelity function (e.g. a computational experiment with a CFD code) and an expensive high fidelity function (e.g. a wind tunnel experiment) to generate a data sample in order to construct a regression model of a high fidelity function. The key question in this setting is how the sizes of the high and low fidelity data samples should be selected in order to stay within a given computational budget and maximize accuracy of the regression model prior to committing resources on data acquisition. In this presentation we discuss minimax interpolation errors for single and variable fidelity scenarios for a multivariate Gaussian process regression. Evaluation of the minimax errors allows us to identify cases when the variable fidelity data provides better interpolation accuracy than the exclusively high fidelity data for the same computational budget. These results allow us to calculate the optimal shares of variable fidelity data samples under the given computational budget constraint. Real and synthetic data experiments suggest that using the obtained optimal shares often outperforms natural heuristics in terms of the regression accuracy.
Based on papers:
01.11.2016 |
 

 

© Institute for Information Transmission Problems of the Russian Academy of Sciences (Kharkevich Institute), 2024
About  |  Contacts  |  Ïðîòèâîäåéñòâèå êîððóïöèè