18177163. BOOSTING GRAPH EXPLAINABILITY USING SELF-GUIDED DATA AUGMENTATION simplified abstract (Robert Bosch GmbH)
BOOSTING GRAPH EXPLAINABILITY USING SELF-GUIDED DATA AUGMENTATION
Organization Name
Inventor(s)
Piyush Chawla of Columbus OH (US)
Thang Doan of Santa Clara CA (US)
BOOSTING GRAPH EXPLAINABILITY USING SELF-GUIDED DATA AUGMENTATION - A simplified explanation of the abstract
This abstract first appeared for US patent application 18177163 titled 'BOOSTING GRAPH EXPLAINABILITY USING SELF-GUIDED DATA AUGMENTATION
Simplified Explanation: The patent application describes a method for training a model to determine graph similarity by receiving two graphs as inputs, aligning nodes between the graphs, and refining the model based on training losses.
Key Features and Innovation:
- Training a model to determine graph similarity
- Aligning nodes between two graphs
- Modifying graphs based on aligned nodes
- Refining the model based on training losses
Potential Applications: This technology can be applied in various fields such as:
- Bioinformatics
- Social network analysis
- Image recognition
- Fraud detection
Problems Solved: The technology addresses the following problems:
- Efficient graph similarity determination
- Improved accuracy in aligning nodes
- Enhanced model training for graph analysis
Benefits: The benefits of this technology include:
- Increased accuracy in graph comparison
- Faster model training process
- Enhanced performance in various applications
Commercial Applications: Title: Graph Similarity Model Training Technology for Enhanced Data Analysis This technology can be utilized in commercial applications such as:
- Data analytics software
- Recommendation systems
- Network security tools
Prior Art: Readers can explore prior art related to graph similarity models, node alignment techniques, and graph analysis methods in academic journals and patent databases.
Frequently Updated Research: Stay updated on the latest advancements in graph similarity models, node alignment algorithms, and graph analysis techniques to enhance the application of this technology.
Questions about Graph Similarity Model Training Technology 1. How does this technology improve graph analysis processes? 2. What are the potential limitations of using this method for training graph similarity models?
Original Abstract Submitted
A method for training a model for determining graph similarity is disclosed. The method comprises receiving a first graph and a second graph as training inputs, the first graph and the second graph each including nodes connected by edges. The method further comprises applying a model to the first graph and the second graph to determine (i) pairs of aligned nodes between the first graph and the second graph and (ii) a first training loss. The method further comprises generating a first augmented graph by modifying the first graph depending on the pairs of aligned nodes. The method further comprises applying the model to the first graph and the first augmented graph to determine a second training loss. The method further comprises refining the model based on the first training loss and the second training loss.
