It’s been said in the AI community that neural nets are always the ‘second best’ solution. Meaning, that a well-groomed expert system will always beat them.
That’s interesting, the book I’m reading now claims just the opposite. In “Super Crunchers: How Thinking by Numbers is the new way to be Smart” by Ayres he describes how algorithms beat human experts in field after field, whether it’s predicting the future success of baseball players, if a particular vintage wine will be good, or how the Chief Justices will vote in a court case. In test after test, the algorithms beat the experts.
Neural-nets are a particular sort of generic problem-solving algorithm. What I’m saying is that a custom algorithm (like those mentioned in the book), tuned to solve a certain problem will always beat the neural net approach.
Neural-nets are a particular sort of generic problem-solving algorithm. What I’m saying is that a custom algorithm (like those mentioned in the book), tuned to solve a certain problem will always beat the neural net approach.
Neural-nets are a particular sort of generic problem-solving algorithm. What I’m saying is that a custom algorithm (like those mentioned in the book), tuned to solve a certain problem will always beat the neural net approach.
Neural-nets are a particular sort of generic problem-solving algorithm. What I’m saying is that a custom algorithm (like those mentioned in the book), tuned to solve a certain problem will always beat the neural net approach.
Neural-nets are a particular sort of generic problem-solving algorithm. What I’m saying is that a custom algorithm (like those mentioned in the book), tuned to solve a certain problem will always beat the neural net approach.
