Navid Ardeshir
I'm a third year Ph.D. student in statistics at Columbia University,
advised by Daniel Hsu
and Arian Maleki.
I earned my bachelors in Electrical Engineering at Sharif University in 2019
which gave me the opportunity to work with my wonderful advisers Kasra Alishahi,
and Amin Gohari.
I am broadly interested in machine learning theory and high dimensional statistics,
particularly studying the behavior of overparameterized models. For more info
take a glance at my papers or my
CV.
I also started to write down some blog posts about interesting mathematical phenomena in statistics and ML!
- Is Lasso Optimal?: A short article on optimality of Lasso estimator
- Exploration Methods Based on Value Function: A domain specific literature review on exploration methods in Reinforcement Learning
- Online Learning Through The Lens of Potential Descent: A note on blackwell approachability and it’s aaplications to expert advice.
- CLT for Empirical Transportation Cost in General Dimensions: A summary note on Del Bario’s paper regarding how fast empirical transports converge to an arbitraty measure compared to their population version.
- Boosting From a Theoretical Point of View: Slides on boosting from a game theoretical perspective.