Marife Defante

Title: Marife Defante: Innovator in Machine Learning

Introduction

Marife Defante is a notable inventor based in Woodinville, WA (US). He has made significant contributions to the field of machine learning, particularly through his innovative patent. His work exemplifies the intersection of technology and creativity, showcasing how inventive minds can shape the future.

Latest Patents

Marife Defante holds a patent titled "Modelling causation in machine learning." This patent describes a method that involves sampling a temporal causal graph from a temporal graph distribution. It specifies probabilities of directed causal edges between different variables of a feature vector at a present time step, as well as from one variable at a preceding time step to another variable at the present time step. The method identifies a present parent, which is a cause of the selected variable in the present time step, and a preceding parent, which is a cause of the selected variable from the preceding time step. The process includes inputting values of each identified present and preceding parent into respective encoders, resulting in embeddings. These embeddings are then combined and input into a decoder to reconstruct the value of the selected variable. Marife has 1 patent to his name.

Career Highlights

Marife Defante is currently employed at Microsoft Technology Licensing, LLC. His role at this prominent technology company allows him to work on cutting-edge innovations in machine learning and artificial intelligence. His expertise and contributions are invaluable to the ongoing advancements in these fields.

Collaborations

Marife has collaborated with talented individuals such as Wenbo Gong and Cheng Zhang. These partnerships highlight the collaborative nature of innovation, where diverse expertise can lead to groundbreaking developments.

Conclusion

Marife Defante is a distinguished inventor whose work in machine learning is paving the way for future advancements. His patent reflects a deep understanding of causal relationships in data, which is crucial for the evolution of intelligent systems.