Learning to Optimize via Dual space Preconditioning

Sélim Chraibi, Adil Salim, Samuel Horváth, Filip Hanzely, Peter Richtárik

Abstract: Preconditioning an minimization algorithm improve its convergence and can lead to a minimizer in one iteration in some extreme cases. There is currently no analytical way for finding a suitable preconditioner. We present a general methodology for learning the preconditioner and show that it can lead to dramatic speed-ups over standard optimization techniques.

Code: https://drive.google.com/file/d/10XrwojJQUxUWb2Khp9cFp5JyM7usQt5D/view?usp=sharing

Keywords: Optimization, meta-learning