18313907. NEURAL GRAPH REVEALERS simplified abstract (Microsoft Technology Licensing, LLC)
Contents
NEURAL GRAPH REVEALERS
Organization Name
Microsoft Technology Licensing, LLC
Inventor(s)
Harsh Shrivastava of Redmond WA (US)
Urszula Stefania Chajewska of Camano Island WA (US)
NEURAL GRAPH REVEALERS - A simplified explanation of the abstract
This abstract first appeared for US patent application 18313907 titled 'NEURAL GRAPH REVEALERS
The present disclosure involves recovering a sparse feature graph from input data containing samples and features using a fully connected neural network. The neural network learns a regression of the input data and identifies direct connections between features while satisfying sparsity constraints. This regression helps in reconstructing a feature graph showing connections between input data features, which can be interactively presented for user exploration.
- Fully connected neural network utilized
- Regression of input data learned
- Direct connections between features identified
- Sparsity constraints satisfied
- Feature graph reconstructed
- Interactive presentation for user exploration
Potential Applications: - Data analysis and visualization - Pattern recognition - Machine learning algorithms - Network analysis
Problems Solved: - Efficiently recovering sparse feature graphs - Identifying direct connections between features - Enhancing user understanding of input data
Benefits: - Improved data interpretation - Enhanced feature relationship visualization - Facilitates insights discovery - Optimizes machine learning processes
Commercial Applications: Title: "Enhanced Data Visualization and Analysis Tool for Machine Learning" This technology can be applied in industries such as: - Data analytics - Artificial intelligence - Financial modeling - Healthcare diagnostics
Questions about the Technology: 1. How does the fully connected neural network contribute to the recovery of sparse feature graphs? 2. What are the potential real-world applications of interactive feature graph presentations?
Frequently Updated Research: Stay updated on advancements in neural network algorithms for feature graph reconstruction and visualization techniques.
Original Abstract Submitted
The present disclosure relates to recovering a sparse feature graph based on input data having a collection of samples and associated features. In particular, the systems described herein utilize a fully connected neural network to learn a regression of the input data and determine direct connections between features of the input data while the neural network satisfies one or more sparsity constraints. This regression may be used to recover a feature graph indicating direct connections between the features of the input data. In addition, the feature graph may be presented via an interactive presentation that enables a user to navigate nodes and edges of the graph to gain insights of the input data and associated features.
