20240006052. SYSTEM IMPLEMENTING GENERATIVE ADVERSARIAL NETWORK ADAPTED TO PREDICTION IN BEHAVIORAL AND/OR PHYSIOLOGICAL CONTEXTS simplified abstract (CORNELL UNIVERSITY)
Contents
- 1 SYSTEM IMPLEMENTING GENERATIVE ADVERSARIAL NETWORK ADAPTED TO PREDICTION IN BEHAVIORAL AND/OR PHYSIOLOGICAL CONTEXTS
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 SYSTEM IMPLEMENTING GENERATIVE ADVERSARIAL NETWORK ADAPTED TO PREDICTION IN BEHAVIORAL AND/OR PHYSIOLOGICAL CONTEXTS - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Potential Applications
- 1.6 Problems Solved
- 1.7 Benefits
- 1.8 Original Abstract Submitted
SYSTEM IMPLEMENTING GENERATIVE ADVERSARIAL NETWORK ADAPTED TO PREDICTION IN BEHAVIORAL AND/OR PHYSIOLOGICAL CONTEXTS
Organization Name
Inventor(s)
Daniel A. Adler of Yardley PA (US)
Tanzeem Choudhury of Pittsford NY (US)
Vincent W.-S. Tseng of Boston MA (US)
SYSTEM IMPLEMENTING GENERATIVE ADVERSARIAL NETWORK ADAPTED TO PREDICTION IN BEHAVIORAL AND/OR PHYSIOLOGICAL CONTEXTS - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240006052 titled 'SYSTEM IMPLEMENTING GENERATIVE ADVERSARIAL NETWORK ADAPTED TO PREDICTION IN BEHAVIORAL AND/OR PHYSIOLOGICAL CONTEXTS
Simplified Explanation
The abstract describes a method that uses a generative adversarial network to predict changes in behavior and physiology of a subject based on data collected over time. The network is designed to perform multi-task learning across multiple subjects, treating changes in different features as separate but linked tasks. The network includes separate discriminators for each feature and for different clusters of subjects, and combines their outputs to generate predictions.
- The method uses data to predict changes in behavior and physiology of a subject.
- A generative adversarial network is employed, which is a type of machine learning model.
- The network is trained to perform multi-task learning, treating changes in different features as separate but linked tasks.
- Separate discriminators are used for each feature and for different clusters of subjects.
- The network combines the outputs of the discriminators to generate predictions.
- Automated remedial actions can be taken based on the generated predictions.
Potential Applications
- Healthcare: Predicting changes in patient behavior and physiology to enable proactive interventions.
- Sports: Anticipating changes in athlete performance and health to optimize training and prevent injuries.
- Finance: Forecasting changes in market behavior and economic indicators to inform investment strategies.
- Customer Behavior: Predicting changes in consumer preferences and buying patterns to personalize marketing efforts.
Problems Solved
- Early detection: The method allows for the early prediction of changes in behavior and physiology, enabling timely interventions.
- Proactive actions: By generating predictions, automated remedial actions can be taken to address potential issues before they escalate.
- Multi-task learning: The approach treats changes in different features as separate tasks, improving the accuracy of predictions.
Benefits
- Improved outcomes: Early detection and proactive actions can lead to better health outcomes, performance, and decision-making.
- Personalization: The method can tailor interventions and strategies based on individual subjects and their specific needs.
- Efficiency: Automated remedial actions reduce the need for manual intervention and can save time and resources.
Original Abstract Submitted
a method comprises obtaining data characterizing a given subject over time, applying at least a portion of the obtained data to a generative adversarial network adapted to generate a prediction of at least one change in at least one of behavior and physiology of the given subject from the obtained data, and executing at least one automated remedial action relating to the given subject based at least in part on the generated prediction. the generative adversarial network is configured to implement multi-task learning, across a plurality of subjects, in which changes in multiple distinct features are treated as separate but linked tasks. the generative adversarial network comprises separate discriminators for each of the multiple distinct features and separate discriminators for each of a plurality of different clusters of respective subsets of the plurality of subjects, and combines outputs of respective ones of the discriminators for the features and the clusters in generating the prediction.
