That's a good question but I think to make a good prediction is one thing and to make a prediction about a football game that is guaranteed to work is another thing. As he mentions in the article, it is really the same old combinatorial explosion and curse of dimensionality at work. No matter how much data you've got reality is ready with things not foreseen. I think everybody pretty much agrees on that in so much as an AI is not going to have mystical powers which I would say figuratively would exceed "God himself" type powers.
What is going to happen next in the world? How can we ever be assured? https://products.kitsapsun.com/archive/1999/11-17/0146_the_apple_cup___82_upset_was_bigg.html On 8/8/18, Dorian Aur via AGI <[email protected]> wrote: > How do you know what God can or can't do? > >> >> God himself cannot predict the outcome of football. >>> >>> On 8/8/18, Matt Mahoney via AGI <[email protected]> wrote: >>> > Nobody knows how to test for an adult IQ of 250. What is the IQ of >>> > your >>> > smart phone? >>> > >>> > Copies of a simulation give you copies of the same prediction. Markets >>> beat >>> > individuals because individuals are different. The PAQ compression >>> > algorithm works on this principle. It averages together many >>> > independent >>> > models to predict the next input bit. >>> > >>> > >>> > >>> > On Tue, Aug 7, 2018, 10:16 PM MP via AGI <[email protected]> wrote: >>> > >>> >> Or you could have one AGI with the mental capacity to simulate a mind >>> >> with >>> >> an IQ of 250, then run simulations of those millions of minds one >>> >> right >>> >> after the other in a serial fashion. I’m sure it’d be smart enough to >>> >> figure out when to stop and make a descision based on the minds it >>> >> serially >>> >> simulates. >>> >> >>> >> My point is, minds beat math. >>> >> >>> >> Sent from ProtonMail Mobile >>> >> >>> >> >>> >> On Tue, Aug 7, 2018 at 9:11 PM, Matt Mahoney via AGI >>> >> <[email protected]> >>> >> wrote: >>> >> >>> >> Prediction in sports is big business. Same as betting on stocks and >>> other >>> >> investments. The best predictor is the market price. If there was an >>> >> algorithm that could do better then everyone would use it. You are >>> going >>> >> to >>> >> have equal amounts of wins and losses no matter what. >>> >> >>> >> For AGI to beat the markets it has to be smarter than not just one >>> human >>> >> but millions of humans. >>> >> >>> >> On Tue, Aug 7, 2018, 5:29 PM MP via AGI <[email protected]> wrote: >>> >> >>> >>> The issue lies in trying to use simple mathematical structures to >>> >>> predict >>> >>> the behavior of the human mind. >>> >>> >>> >>> You need an AGI with the irrationality and flexibility of the human >>> mind >>> >>> to predict their behaviors. >>> >>> >>> >>> Sent from ProtonMail Mobile >>> >>> >>> >>> >>> >>> On Tue, Aug 7, 2018 at 10:48 AM, Dorian Aur via AGI >>> >>> <[email protected]> >>> >>> wrote: >>> >>> >>> >>> Predictive algorithms are no better at telling the future, see >>> >>> >>> >>> >>> >>> https://medium.com/@bakhtiyari/artificial-intelligence- >>> failed-in-world-cup-2018-6af10602206a >>> >>> >>> >>> Where is the issue? >>> >>> >>> >>> Dorian >>> >>> >>> >>> *Artificial General Intelligence List <https://agi.topicbox.com/late >>> st>* >>> >> / AGI / see discussions <https://agi.topicbox.com/groups/agi> + >>> >> participants <https://agi.topicbox.com/groups/agi/members> + delivery >>> >> options <https://agi.topicbox.com/groups/agi/subscription> Permalink >>> >> <https://agi.topicbox.com/groups/agi/T63cec70f214bbcf8-M72ba >> >> ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T63cec70f214bbcf8-M89911d014731e4d8b1b72353 Delivery options: https://agi.topicbox.com/groups/agi/subscription
