Unlocking the Hidden Structure of Data: Research by Professor Yusu Wang
Professor Yusu Wang from UC San Diego is improving artificial intelligence systems by revealing hidden structures in complex data using geometry and topology.
UC San Diego Professor Yusu Wang utilizes geometry and topology to uncover hidden structures within complex data. This approach enables machine learning systems to capture relationships that traditional methods might miss.
The Core Logic of Topology
Topologists state that despite differences in outward appearance, certain objects share a fundamental mathematical shape. This approach allows focusing on the truly essential structure by eliminating unnecessary details.
Yusu Wang notes that this seemingly simple mathematical idea has guided her throughout her career and forms the basis of her complex data analysis work.
Artificial Intelligence and Data Analysis
Wang's research spans a wide area from geometric and topological data analysis to geometric deep learning, graph learning, and neural algorithmic reasoning.
These studies find practical applications in various disciplines such as neuroscience, materials science, and particle physics.
TILOS and Interdisciplinary Studies
Yusu Wang also serves as the director of the Institute for Learning-enabled Optimization at Scale (TILOS). The institute brings together artificial intelligence and optimization.
This combination aims to solve problems in computationally challenging fields such as chip design, robotics, and communication networks.