Improving Data Understanding with Graphical Hypergeometric Networks
summary
The talk introduces graphical hypergeometric networks (HNet), a method that tests associations between variables for statistical significance using hypergeometric-based inference. HNet is designed to learn associations from datasets with mixed data types and where the functional relationships are unknown, improving the data-understanding step in real-world data analysis.
AI-written summary · about
chapters (11)
- 0:00Welcome
- 1:36Data preprocessing challenges
- 4:13Hypergeometric Networks benefits
- 7:25Underused in practice
- 10:16Building step by step
- 13:24Tool demo
- 17:26Label relationships insights
- 21:01Applications and case studies
- 25:02Model deployment
- 27:55Conclusion and future
- 30:06Q&A
related talks
01h 01mToward Universal Graph Representations: Foundations and Frontiers of Graph Foundation Models
32mNew Approaches To Learning Nonparametric (Latent) Causal Graphical Models
39mUsing Graph Neural Networks to Embrace the Dependency Within Your Data
59mTransforming DataFrames: Intro to Graphs & GNNs with PyTorch Geometric
01h 23mProbabilistic Graphical Models, HMMs using PGMPY
01h 41mGraph Intelligence: Enhance Reasoning and Retrieval Using Graph Analytics
16mPreview: Graph NNs
01h 00mTensor Graphical Models For Complex And High Dimensional Data