Bayesian methods have been a popular technique for developing artificial intelligence, e.g., Nate Silver claimed a success of utilization of Bayesian methods in predicting the US election. I wonder if Bayesian has the real potential to challenge the real human intelligence. Is there any other counterpart philosophy which is more probable to be able to challenge the real human intelligence?
I think that Bayesian methodology has not potential to challenge real human intelligence at all.
The Bayesian methodology is popular because has been used succesfully to analyzing, recognition and prediction in numerous subjects. But that success is based in processing of big data. In other words, Bayesian results are based in statistics no in a data comprehension as humans does. There are not a conscious knowledge. Human intelligence not only proceed of analysis of statistics data, we use also logic and intuition to deal with practical problems. Yet more, human intelligence differs of artificial intelligence by their capability to reflect about our own thinking. As far as I know, until now have failed all attempts to create reflexive thinking on machines.
We can not say its not potential ..it again depends on your implementation .....anything close to analysis ....bayesian is not a good option ...but leave a good impact on automation systems.
Thanks Carmen and Alokkumar for your interests in answering this question. So Carmen, as you said "As far as I know, until now have failed all attempts to create reflexive thinking on machines", could you pls provide any references on your mentioned "attempts to create reflexive thinking on machines" ?
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Artificial Intelligence