System and method for vigorous artificial intelligence

Abstract

A system and method for predicting a characteristic of an object in an artificial intelligence system. The method includes evaluating the object using a first model to produce a first prediction of a characteristic of the object. The object is evaluated using a second model to produce a second prediction of the characteristic of the object, the second model being dissimilar to the first model. A final prediction of the characteristic of the object is generated as a function of dynamic weightings of the first prediction and the second prediction.

1 . A method of characterizing an object input to an artificial intelligence (AI) system, comprising:
 evaluating said object using a first model to produce a first prediction of a characteristic of said object;   evaluating said object using a second model to produce a second prediction of said characteristic of said object, said second model being dissimilar to said first model; and   generating a final prediction of said characteristic of said object as a function of dynamic weightings of said first prediction and said second prediction.   
     
     
         2 . The method recited in  claim 1 , wherein evaluating said object using said first or second model further comprises determining a quality of said first or second prediction, respectively. 
     
     
         3 . The method recited in  claim 2 , wherein said quality comprises a measure of the confidence in said prediction. 
     
     
         4 . The method recited in  claim 2 , wherein said dynamic weightings are a function of said quality. 
     
     
         5 . The method recited in  claim 1 , wherein said dynamic weightings are a function of at least one external input. 
     
     
         6 . The method recited in  claim 1 , wherein said dynamic weightings are a function of at least one predefined rule. 
     
     
         7 . The method recited in  claim 1 , wherein said evaluating said object using first and second models are executed in parallel. 
     
     
         8 . The method recited in  claim 1 , wherein said first model comprises a neural network. 
     
     
         9 . The method recited in  claim 8 , further comprising training said first model using a corpus of data. 
     
     
         10 . The method recited in  claim 1 , wherein one of said first and second models comprises a Fast Fourier Transform. 
     
     
         11 . An artificial intelligence (AI) system for characterizing an input object, comprising:
 at least one processor; and,   at least one memory, said at least one memory containing instructions which, when executed by said at least one processor, are operative to:
 evaluate said object using a first model to produce a first prediction of a characteristic of said object; 
 evaluate said object using a second model to produce a second prediction of said characteristic of said object, said second model being dissimilar to said first model; and, 
 generate a final prediction of said characteristic of said object as a function of dynamic weightings of said first prediction and said second prediction. 
   
     
     
         12 . The AI system recited in  claim 11 , wherein evaluating said object using said first or second model further comprises determining a quality of said first or second prediction, respectively. 
     
     
         13 . The AI system recited in  claim 12 , wherein said quality comprises a measure of the confidence in said prediction. 
     
     
         14 . The AI system recited in  claim 12 , wherein said dynamic weightings are a function of said quality. 
     
     
         15 . The AI system recited in  claim 11 , wherein said dynamic weightings are a function of at least one external input. 
     
     
         16 . The AI system recited in  claim 11 , wherein said dynamic weightings are a function of at least one predefined rule. 
     
     
         17 . The AI system recited in  claim 11 , wherein the operations of evaluating said object using said first model and evaluating said object using said second model are executed in parallel. 
     
     
         18 . The AI system recited in  claim 11 , wherein said first model comprises a neural network. 
     
     
         19 . The AI system recited in  claim 18 , further comprising the operation of training said first model using a corpus of data. 
     
     
         20 . The AI system recited in  claim 11 , wherein one of said first and second models comprises a Fast Fourier Transform. 
     
     
         21 . An artificial intelligence (AI) modulator for characterizing an object, comprising:
 a processor; and,   a memory, said memory containing instructions which, when executed by said processor, are operative to cause said AI modulator to:
 receive a first evaluation of said object from a first model, said first evaluation comprising a first prediction of a characteristic of said object; 
 receive a second evaluation of said object from a second model, said second evaluation comprising a second prediction of said characteristic of said object, said second model being dissimilar to said first model; and, 
 generate a final prediction of said characteristic of said object as a function of dynamic weightings of said first prediction and said second prediction. 
   
     
     
         22 . The AI modulator recited in  claim 21 , wherein said first or second evalutions of said object further comprises a quality of said first or second prediction, respectively. 
     
     
         23 . The AI modulator recited in  claim 22 , wherein said quality comprises a measure of the confidence in said prediction. 
     
     
         24 . The AI modulator recited in  claim 22 , wherein said dynamic weightings are a function of said quality. 
     
     
         25 . The AI modulator recited in  claim 21 , wherein said dynamic weightings are a function of at least one external input. 
     
     
         26 . The AI modulator recited in  claim 21 , wherein said dynamic weightings are a function of at least one predefined rule. 
     
     
         27 . The AI modulator recited in  claim 21 , wherein one of said first and second models comprises a neural network. 
     
     
         28 . The AI modulator recited in  claim 21 , wherein one of said first and second models comprises a Fast Fourier Transform.