Melih Kandemir

A new method enhances predictions in complex dynamical systems.

PAC algorithm improves exploration-exploitation balance in reinforcement learning.

MOMBO enhances decision-making in offline reinforcement learning with improved uncertainty estimation.