18539022. CONCURRENT OPTIMIZATION OF MACHINE LEARNING MODEL PERFORMANCE simplified abstract (QUALCOMM Incorporated)
Contents
- 1 CONCURRENT OPTIMIZATION OF MACHINE LEARNING MODEL PERFORMANCE
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 CONCURRENT OPTIMIZATION OF MACHINE LEARNING MODEL PERFORMANCE - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Potential Applications
- 1.6 Problems Solved
- 1.7 Benefits
- 1.8 Potential Commercial Applications
- 1.9 Possible Prior Art
- 1.10 Original Abstract Submitted
CONCURRENT OPTIMIZATION OF MACHINE LEARNING MODEL PERFORMANCE
Organization Name
Inventor(s)
Serag Gadelrab of Markham (CA)
James Lyall Esliger of Tillsonburg (CA)
Meghal Varia of North York (CA)
Alwyn Dos Remedios of Maple (CA)
CONCURRENT OPTIMIZATION OF MACHINE LEARNING MODEL PERFORMANCE - A simplified explanation of the abstract
This abstract first appeared for US patent application 18539022 titled 'CONCURRENT OPTIMIZATION OF MACHINE LEARNING MODEL PERFORMANCE
Simplified Explanation
Certain aspects of the present disclosure provide techniques for concurrently performing inferences using a machine learning model and optimizing parameters used in executing the machine learning model. An example method generally includes receiving a request to perform inferences on a data set using the machine learning model and performance metric targets for performance of the inferences. At least a first inference is performed on the data set using the machine learning model to meet a latency specified for generation of the first inference from receipt of the request. While performing the at least the first inference, operational parameters resulting in inference performance approaching the performance metric targets are identified based on the machine learning model and operational properties of the computing device. The identified operational parameters are applied to performance of subsequent inferences using the machine learning model.
- Techniques for concurrently performing inferences and optimizing parameters using a machine learning model.
- Method involves receiving a request for inferences on a data set with performance metric targets.
- First inference is performed to meet specified latency, while identifying operational parameters for optimal performance.
- Operational parameters are applied to subsequent inferences using the machine learning model.
Potential Applications
This technology could be applied in various fields such as:
- Healthcare for optimizing medical diagnosis processes.
- Finance for improving fraud detection systems.
- Autonomous vehicles for enhancing real-time decision-making capabilities.
Problems Solved
This technology helps in:
- Improving the efficiency and accuracy of machine learning models.
- Reducing latency in generating inferences.
- Optimizing operational parameters for better performance.
Benefits
The benefits of this technology include:
- Enhanced performance of machine learning models.
- Faster generation of inferences.
- Improved overall system efficiency.
Potential Commercial Applications
Optimizing Machine Learning Model Parameters for Enhanced Performance
Possible Prior Art
One possible prior art for this technology could be the use of grid search or random search techniques for hyperparameter optimization in machine learning models.
What are the potential limitations of this technology in real-world applications?
The potential limitations of this technology in real-world applications could include:
- Complexity in identifying the optimal operational parameters.
- Dependency on the accuracy of the machine learning model.
- Resource-intensive computational requirements.
How does this technology compare to existing methods for optimizing machine learning model parameters?
This technology stands out from existing methods by:
- Simultaneously performing inferences and optimizing parameters.
- Adapting operational parameters based on real-time performance metrics.
- Enhancing overall efficiency and accuracy of machine learning models.
Original Abstract Submitted
Certain aspects of the present disclosure provide techniques for concurrently performing inferences using a machine learning model and optimizing parameters used in executing the machine learning model. An example method generally includes receiving a request to perform inferences on a data set using the machine learning model and performance metric targets for performance of the inferences. At least a first inference is performed on the data set using the machine learning model to meet a latency specified for generation of the first inference from receipt of the request. While performing the at least the first inference, operational parameters resulting in inference performance approaching the performance metric targets are identified based on the machine learning model and operational properties of the computing device. The identified operational parameters are applied to performance of subsequent inferences using the machine learning model.
