View source for 18052297. SYSTEM AND METHOD FOR TORQUE-BASED STRUCTURED PRUNING FOR DEEP NEURAL NETWORKS simplified abstract (SAMSUNG ELECTRONICS CO., LTD.)

=SYSTEM AND METHOD FOR TORQUE-BASED STRUCTURED PRUNING FOR DEEP NEURAL NETWORKS=


==Organization Name==

[[:Category:SAMSUNG ELECTRONICS CO., LTD.|SAMSUNG ELECTRONICS CO., LTD.]]


[[Category:SAMSUNG ELECTRONICS CO., LTD.]]

==Inventor(s)==

[[:Category:Tien C. Bau of Irvine CA (US)|Tien C. Bau of Irvine CA (US)]][[Category:Tien C. Bau of Irvine CA (US)]]

[[:Category:Arshita Gupta of Irvine CA (US)|Arshita Gupta of Irvine CA (US)]][[Category:Arshita Gupta of Irvine CA (US)]]

[[:Category:Hrishikesh Deepak Garud of Mountain View CA (US)|Hrishikesh Deepak Garud of Mountain View CA (US)]][[Category:Hrishikesh Deepak Garud of Mountain View CA (US)]]

==SYSTEM AND METHOD FOR TORQUE-BASED STRUCTURED PRUNING FOR DEEP NEURAL NETWORKS - A simplified explanation of the abstract==


This abstract first appeared for US patent application 18052297 titled 'SYSTEM AND METHOD FOR TORQUE-BASED STRUCTURED PRUNING FOR DEEP NEURAL NETWORKS

==Simplified Explanation==

The abstract describes a method that involves using a machine learning model trained with a torque-based constraint. The method includes accessing the model, receiving an input, providing the input to the model, receiving an output, and instructing actions based on the output. 

* The machine learning model is trained using a torque-based constraint.
* The method involves receiving an input from an input source and providing it to the model.
* The output from the model is received and used to instruct at least one action.
* The training of the model includes applying a torque-based constraint on filters.
* The first set of filters is adjusted to have a higher concentration of weights than the second set.
* At least one channel of the model is pruned based on the average weight for that channel.

==Potential Applications==
* This method can be applied in various fields where machine learning models are used, such as robotics, autonomous vehicles, and industrial automation.
* It can be used for tasks that require precise control and decision-making based on input data.

==Problems Solved==
* The torque-based constraint helps in training the machine learning model to make more accurate predictions or decisions.
* The adjustment of filters and pruning of channels based on weights can improve the efficiency and performance of the model.

==Benefits==
* By using a torque-based constraint, the machine learning model can be trained to better handle situations that involve torque-related factors.
* The higher concentration of weights in the adjusted filters can enhance the model's ability to capture important features or patterns in the input data.
* Pruning channels based on average weight can reduce the complexity of the model and improve its computational efficiency.


==Original Abstract Submitted==

A method includes accessing a machine learning model, the machine learning model trained using a torque-based constraint. The method also includes receiving an input from an input source and providing the input to the machine learning model. The method also includes receiving an output from the machine learning model. The method also includes instructing at least one action based on the output from the machine learning model. Training the machine learning model includes applying a torque-based constraint on one or more filters of the machine learning model, adjusting, based on applying the torque-based constraint, a first set of one or more filters of the machine learning model to have a higher concentration of weights than a second set of one or more filters of the machine learning model, and pruning at least one channel of the machine learning model based on an average weight for the at least one channel.



[[Category:G06N3/082]]