View source for 17541823. ACTIVE LEARNING DRIFT ANALYSIS AND TRAINING simplified abstract (INTERNATIONAL BUSINESS MACHINES CORPORATION)

=ACTIVE LEARNING DRIFT ANALYSIS AND TRAINING=


==Organization Name==

[[:Category:INTERNATIONAL BUSINESS MACHINES CORPORATION|INTERNATIONAL BUSINESS MACHINES CORPORATION]]


[[Category:INTERNATIONAL BUSINESS MACHINES CORPORATION]]

==Inventor(s)==

[[:Category:Amadou Ba of Navan (IE)|Amadou Ba of Navan (IE)]][[Category:Amadou Ba of Navan (IE)]]

[[:Category:Venkata Sitaramagiridharganesh Ganapavarapu of Elmsford NY (US)|Venkata Sitaramagiridharganesh Ganapavarapu of Elmsford NY (US)]][[Category:Venkata Sitaramagiridharganesh Ganapavarapu of Elmsford NY (US)]]

[[:Category:Seshu Tirupathi of Dublin (IE)|Seshu Tirupathi of Dublin (IE)]][[Category:Seshu Tirupathi of Dublin (IE)]]

[[:Category:Bradley Eck of Dublin (IE)|Bradley Eck of Dublin (IE)]][[Category:Bradley Eck of Dublin (IE)]]

==ACTIVE LEARNING DRIFT ANALYSIS AND TRAINING - A simplified explanation of the abstract==


This abstract first appeared for US patent application 17541823 titled 'ACTIVE LEARNING DRIFT ANALYSIS AND TRAINING

==Simplified Explanation==

The patent application describes a system and method for training a learning model based on determined drift. The system includes a memory and a processor that execute computer executable components stored in the memory. The components include a selection component that selects an ensemble of deep learning regressors, an identification component that identifies drift among the ensemble, an analysis component that analyzes uncertainty samplings to determine when drift occurred, and a training component that trains deep learning models based on the identified drift.

* The system selects an ensemble of deep learning regressors.
* It identifies drift among the ensemble.
* It analyzes uncertainty samplings to determine when drift occurred.
* It trains deep learning models based on the identified drift.

== Potential Applications ==
* This technology can be applied in various fields where deep learning models are used, such as image recognition, natural language processing, and speech recognition.
* It can be used in autonomous vehicles to continuously train the learning model based on drift, improving their ability to adapt to changing road conditions.
* It can be utilized in predictive maintenance systems to detect drift in sensor data and update the learning model accordingly, improving the accuracy of failure predictions.

== Problems Solved ==
* Drift is a common issue in machine learning models, where the model's performance deteriorates over time due to changes in the data distribution.
* This technology solves the problem of detecting and addressing drift in deep learning models, ensuring that the models remain accurate and reliable over time.

== Benefits ==
* By training deep learning models based on identified drift, the system can improve the performance and accuracy of the models.
* The system allows for continuous adaptation and learning, ensuring that the models stay up-to-date with changing data distributions.
* It reduces the need for manual intervention and monitoring, as the system automatically detects and addresses drift in the learning models.


==Original Abstract Submitted==

One or more systems, devices, computer program products and/or computer-implemented methods of use provided herein relate to training a learning model based on determined drift. A system can comprise a memory that stores computer executable components, and a processor that executes the computer executable components stored in the memory, wherein the computer executable components can comprise a selection component that can select an ensemble of deep learning regressors, and an identification component that can identify drift among the ensemble. An analysis component can analyze uncertainty samplings from the ensemble to determine a time instant when drift occurred. A training component can train one or more deep learning models, such as of the deep learning regressors, based upon the identified drift.



[[Category:G06K9/62]]
[[Category:G06N3/04]]