Wilhelmina Joseline Donkoh

This researcher has made significant contributions to advancing machine learning through innovative approaches that combine optimization algorithms with neural network architectures. Their work emphasizes the development of efficient optimizers for deep learning tasks, including those inspired by Adam and AdaGrad. Additionally, they have explored novel model architectures that push the boundaries of how models can process sequential data, as seen in the application of Transformer-based models to various domains. Their research bridges foundational optimization studies with modern architectural innovations, providing new insights into both classical and contemporary techniques in deep learning.