View source for 17548182. COMPUTER OPTIMIZATION OF TASK PERFORMANCE THROUGH DYNAMIC SENSING simplified abstract (INTERNATIONAL BUSINESS MACHINES CORPORATION)

=COMPUTER OPTIMIZATION OF TASK PERFORMANCE THROUGH DYNAMIC SENSING=


==Organization Name==

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


[[Category:INTERNATIONAL BUSINESS MACHINES CORPORATION]]

==Inventor(s)==

[[:Category:Jenny S. Li of Franklinton NC (US)|Jenny S. Li of Franklinton NC (US)]][[Category:Jenny S. Li of Franklinton NC (US)]]

[[:Category:Nirmit V. Desai of Yorktown Heights NY (US)|Nirmit V. Desai of Yorktown Heights NY (US)]][[Category:Nirmit V. Desai of Yorktown Heights NY (US)]]

[[:Category:Dhiraj Joshi of Edison NJ (US)|Dhiraj Joshi of Edison NJ (US)]][[Category:Dhiraj Joshi of Edison NJ (US)]]

[[:Category:Raghu Ramaswamy of Bangalore (IN)|Raghu Ramaswamy of Bangalore (IN)]][[Category:Raghu Ramaswamy of Bangalore (IN)]]

[[:Category:Satish Rajani of Dewas (IN)|Satish Rajani of Dewas (IN)]][[Category:Satish Rajani of Dewas (IN)]]

==COMPUTER OPTIMIZATION OF TASK PERFORMANCE THROUGH DYNAMIC SENSING - A simplified explanation of the abstract==


This abstract first appeared for US patent application 17548182 titled 'COMPUTER OPTIMIZATION OF TASK PERFORMANCE THROUGH DYNAMIC SENSING

==Simplified Explanation==

The patent application describes a method, computer program, and system that use sensors of different modalities to generate inferences for a downstream task. Here is a simplified explanation of the abstract:

* The system includes a processor that receives a request for an inference from a group of sensors at a physical location.
* The processor selects sensors of the main modality to provide data to a pipeline, which includes machine learning models for generating the inference.
* The processor applies an outlier detector to the raw data obtained from the main modality sensors.
* If an outlier is detected, the processor automatically engages sensors of a different modality from the group and obtains new raw data.
* The processor then applies the machine learning models to the new raw data to derive the inference.

Potential applications of this technology:

* Environmental monitoring: Using sensors of different modalities to gather data on air quality, temperature, humidity, etc., and generate inferences about the overall environmental conditions.
* Security systems: Integrating sensors such as cameras, motion detectors, and audio sensors to detect anomalies and generate inferences about potential threats.
* Healthcare monitoring: Utilizing sensors of various modalities to collect data on patients' vital signs, movement, and other health-related information to generate inferences about their well-being.

Problems solved by this technology:

* Integration of multiple sensors: The technology allows for the seamless integration of sensors from different modalities, enabling a more comprehensive understanding of the environment or situation.
* Outlier detection: By applying an outlier detector to the raw data, the system can identify unusual or abnormal readings, triggering the engagement of sensors from different modalities to gather additional data.

Benefits of this technology:

* Improved accuracy: By combining data from multiple sensors, the system can generate more accurate inferences for downstream tasks.
* Real-time adaptability: The system automatically engages different sensors when outliers are detected, ensuring that the most relevant and reliable data is used for generating inferences.
* Flexibility: The technology can be applied to various domains and scenarios, allowing for customizable solutions based on specific needs and requirements.


==Original Abstract Submitted==

A method, computer program product, and system include a processor(s) that engages, based on a request for an inference, from a group of sensors of multiple modalities at a physical location, sensor(s) of a main modality to provide data to a pipeline to generate the inference. The pipeline includes one or more machine learning models which generate the inference for a downstream task. The processor(s) obtains raw data from the sensor(s) of the main modality and applies an outlier detector to the raw data. Based on determining that there is an outlier the processor(s) automatically engages sensor(s) of at least one different modality than the main modality from the group of sensors of multiple modalities and obtains new raw data from the sensor(s) of the at least one different modality. The processor(s) applies the one or more machine learning models to the new raw data to derive the inference.



[[Category:G06N5/04]]
[[Category:G06K9/62]]
[[Category:G06N20/00]]