Evaluation of inferences from multiple models trained on similar sensor inputs

Abstract

A computer-implemented method of sensor input processing, implemented by an agricultural platform comprising a processor and a sensor includes receiving sensor input from the sensor; processing the sensor input by multiple machine learning (ML) algorithms, each using a corresponding ML model for generating labels for objects identified in the sensor input; combining labels generated by each ML algorithm to generate a super-imposed labeled sensor input frame; comparing outputs of the ML algorithms to determine similarities or differences; and using results of the comparing for improving an operational characteristic of the sensor input processing.

1 . A computer-implemented method of sensor input processing, implemented by an agricultural platform comprising a processor and a sensor, comprising:
 receiving sensor input from the sensor;   processing the sensor input by multiple machine learning (ML) algorithms, each using a corresponding ML model for generating labels for objects identified in the sensor input;   combining labels generated by each ML algorithm to generate a super-imposed labeled sensor input frame;   comparing outputs of the ML algorithms to determine similarities or differences; and   using results of the comparing for improving an operational characteristic of the sensor input processing.   
     
     
         2 . The method of  claim 1 , wherein the operation characteristic is improved by:
 performing further training of one or more ML models using the super-imposed labeled sensor input frame and/or the similarities of differences in outputs of the ML algorithms.   
     
     
         3 . The method of  claim 1 , wherein the ML models include ML models that are based on different sets of hyperparameters. 
     
     
         4 . The method of  claim 1 , wherein the ML models include ML models that are different versions of a same baseline ML model that has undergone different training. 
     
     
         5 . The method of  claim 1 , wherein the further training is performed based on user feedback on the super-labeled sensor input frame. 
     
     
         6 . The method of  claim 1 , further comprising:
 generating an ML performance metrics based on the comparison of outputs.   
     
     
         7 . The method of  claim 6 , further comprising:
 presenting the ML performance metrics on a user interface.   
     
     
         8 . An apparatus comprising a processor and a sensor disposed on an agricultural platform, wherein the processor is configured to perform a method of sensor input processing, the method comprising:
 receiving sensor input from the sensor;   processing the sensor input by multiple machine learning (ML) algorithms, each using a corresponding ML model for generating labels for objects identified in the sensor input;   combining labels generated by each ML algorithm to generate a super-imposed labeled sensor input frame;   comparing outputs of the ML algorithms to determine similarities or differences; and   using results of the comparing for improving an operational characteristic of the sensor input processing.   
     
     
         9 . The apparatus of  claim 1 , wherein the operation characteristic is improved by:
 performing further training of one or more ML models using the super-imposed labeled sensor input frame and/or the similarities of differences in outputs of the ML algorithms.   
     
     
         10 . The apparatus of  claim 1 , wherein the ML models include ML models that are based on different sets of hyperparameters. 
     
     
         11 . The apparatus of  claim 1 , wherein the ML models include ML models that are different versions of a same baseline ML model that has undergone different training. 
     
     
         12 . The apparatus of  claim 1 , wherein the further training is performed based on user feedback on the super-labeled sensor input frame. 
     
     
         13 . The apparatus of  claim 1 , further comprising:
 generating an ML performance metrics based on the comparison of outputs.   
     
     
         14 . The apparatus of  claim 6 , further comprising:
 presenting the ML performance metrics on a user interface.   
     
     
         15 . A computer-readable medium having code stored thereon, the code, upon execution by a processor, causing the processor to implement a method of sensor input processing, comprising:
 receiving sensor input from a sensor of an agricultural platform;   processing the sensor input by multiple machine learning (ML) algorithms, each using a corresponding ML model for generating labels for objects identified in the sensor input;   combining labels generated by each ML algorithm to generate a super-imposed labeled sensor input frame;   comparing outputs of the ML algorithms to determine similarities or differences; and   using results of the comparing for improving an operational characteristic of the sensor input processing.   
     
     
         16 . The computer-readable medium of  claim 1 , wherein the operation characteristic is improved by:
 performing further training of one or more ML models using the super-imposed labeled sensor input frame and/or the similarities of differences in outputs of the ML algorithms.   
     
     
         17 . The computer-readable medium of  claim 1 , wherein the ML models include ML models that are based on different sets of hyperparameters. 
     
     
         18 . The computer-readable medium of  claim 1 , wherein the ML models include ML models that are different versions of a same baseline ML model that has undergone different training. 
     
     
         19 . The computer-readable medium of  claim 1 , wherein the further training is performed based on user feedback on the super-labeled sensor input frame. 
     
     
         20 . The computer-readable medium of  claim 1 , wherein the method further includes:
 generating an ML performance metrics based on the comparison of outputs, or   presenting the ML performance metrics on a user interface.