Abass Mohammed

The researcher has made significant contributions to the development of efficient optimization algorithms for machine learning, particularly focusing on variance reduction techniques within gradient descent methods. Their work includes novel variants such as SVRG (Stochastic Variance Reduced Gradient) and SAGA (Stochastic Average Gradient Algorithm), which have improved the scalability and performance of these methods across various deep learning applications. The research also extends to constrained optimization problems, demonstrating their applicability in real-world scenarios where algorithmic efficiency is crucial for solving large-scale datasets.