18154269. SELECTING LEARNING MODEL simplified abstract (Telefonaktiebolaget LM Ericsson (publ))
Contents
SELECTING LEARNING MODEL
Organization Name
Telefonaktiebolaget LM Ericsson (publ)
Inventor(s)
Aitor Hernandez Herranz of STOCKHOLM (SE)
[[:Category:José Ara�jo of STOCKHOLM (SE)|José Ara�jo of STOCKHOLM (SE)]][[Category:José Ara�jo of STOCKHOLM (SE)]]
Soma Tayamon of STOCKHOLM (SE)
SELECTING LEARNING MODEL - A simplified explanation of the abstract
This abstract first appeared for US patent application 18154269 titled 'SELECTING LEARNING MODEL
Simplified Explanation
The abstract describes a method for dynamically selecting a learning model for a sensor device. The learning model is used to determine output data based on sensor input. The method involves the following steps:
- Detecting the need for a new learning model based on the performance of the currently loaded learning model in the sensor device.
- Determining feature candidates based on sensor data from different sources.
- Selecting a new learning model from a set of candidate learning models based on the feature candidates and input features of each candidate.
- Triggering the new learning model to be loaded on the sensor device, replacing the currently loaded learning model.
Potential applications of this technology:
- Internet of Things (IoT) devices that rely on sensor data for decision-making.
- Autonomous vehicles that need to adapt their learning models based on changing sensor inputs.
- Smart home devices that optimize their performance based on sensor data.
Problems solved by this technology:
- Outdated or underperforming learning models can be replaced with more suitable models.
- The method allows for continuous improvement and adaptation of the learning model based on changing sensor data.
- It enables efficient utilization of sensor data from different sources to select the most appropriate learning model.
Benefits of this technology:
- Improved accuracy and performance of the sensor device by using the most suitable learning model.
- Adaptability to changing sensor inputs and environmental conditions.
- Efficient utilization of sensor data from different sources to optimize the learning model selection process.
Original Abstract Submitted
According to a first aspect, it is presented a method for dynamically selecting a learning model for a sensor device. The learning model is configured for determining output data based on sensor. The method includes the steps of: detecting a need for a new learning model for the sensor device based on performance of a currently loaded learning model in the sensor device; determining at least one feature candidate based on sensor data from the at least one sensor, wherein each one of the at least one feature candidate is associated with a different source of sensor data; selecting a new learning model, from a set of candidate learning models, based on the at least one feature candidate and input features of each one of the candidate learning models; and triggering the new learning model to be loaded on the sensor device, replacing the currently loaded learning model.
