Aryan Eftekhari

Welome :)

I'm a researcher and lecturer at Università della Svizzera italiana in the Faculty of Informatics, where I also serve as the study coordinator for the Master’s program in Computational Science. I work part-time in collaboration with the University of Lausanne in the Faculty of Business and Economics, where I am developing dynamic multi-resource carbon cycle models.

My research interests focus on multivariate statistical analysis and machine learning for high-dimensional, low-sample-size data, where the complexity of the models and the desired accuracy typically necessitate significantly more data than is available. I develop parallel and structure-exploiting algorithms aimed at enhancing both computational efficiency and model robustness. Some relevant application fields of this work are neuroimaging, genomics, economics, finance, and climate change modeling.

Ongoing projects:

Cloud-Enabled High-Dimensional Low-Sample Size Machine Learning: Sparse Precision Matrix Estimation
Can Economic Policy Mitigate Climate Change?
Frontiers in Data Science and Computing: Bridging Innovation and Education

News & Other

Economists cash in on efficient, high-performance computing methodEconomists have previously made little use of high-performance computers (HPC) in their research. This is despite the fact that the complex interactions and heterogeneity of their models can quickly cause them to reach hundreds of dimensions, which cannot be calculated using conventional methods. In the past, simplified models were therefore often formulated for answering complex questions. These models solved some problems, but they could also provide false predictions, explains Simon Scheidegger, Senior Assistant at the University of Zurich's Department of Banking and Finance. For example, quantitatively studying optimal monetary policy in the wake of a financial crisis cannot be properly achieved using the conventional methods. However, calculating high-dimensional models on a supercomputer is not easy either. Until recently, researchers lacked appropriate numerical analysis and highly efficient software.