Recent Advances in Structural Learning for Probabilistic Graphical Models
Event details
Date | 12.03.2021 |
Hour | 16:15 › 17:30 |
Speaker | Genevera Allen, Rice University |
Location | Online |
Category | Conferences - Seminars |
Probabilistic Graphical Models represent probability distributions as a graph with edges denoting conditional dependence relationships between random variables. These models have been studied and developed in computer science, probability, and statistics, with wide ranging applications in artificial intelligence, computer vision, systems biology, physics, and finance, to name a few.
This talk will review structural learning in probabilistic graphical models, which seeks to learn the unknown edges or conditional dependencies between random variables, and highlight new methods and theory from my research group for graph structural learning from large and complex data. Specifically, I will discuss graph learning for non-Gaussian data including data with extreme events, for mixed data via data integration, for non-simultaneously recorded or non-aligned data via Graph Quilting, and for learning in the presence of latent variables.
Additionally, I will present several applications of these approaches to finance, genomics, and neuroscience.
Probabilistic Graphical Models represent probability distributions as a graph with edges denoting conditional dependence relationships between random variables. These models have been studied and developed in computer science, probability, and statistics, with wide ranging applications in artificial intelligence, computer vision, systems biology, physics, and finance, to name a few.
This talk will review structural learning in probabilistic graphical models, which seeks to learn the unknown edges or conditional dependencies between random variables, and highlight new methods and theory from my research group for graph structural learning from large and complex data. Specifically, I will discuss graph learning for non-Gaussian data including data with extreme events, for mixed data via data integration, for non-simultaneously recorded or non-aligned data via Graph Quilting, and for learning in the presence of latent variables. Additionally, I will present several applications of these approaches to finance, genomics, and neuroscience.
Practical information
- Informed public
- Free
Organizer
- Sofia Olhede
Contact
- Maroussia Schaffner