The "worlds" in which these narrow AI's operate are largely amenable to simulation, which is why they are advancing so rapidly: Just throw more hardware at simulating more such worlds and you can get a really good approximation of "intelligent" behavior in the real world when deployed. Yes, it will lack a lot of nuance, and that's where integrating shared real-world experience becomes important. (Note: This is not "meta-learning" -- that's something else.)
The really interesting stuff is happening in symbolic regression. On Fri, Nov 29, 2019 at 1:06 PM Matt Mahoney <[email protected]> wrote: > Playing Go is narrow AI. Scheduling flights is narrow AI. Driving a truck > is narrow AI. Doing all of these and a million other things is AGI. Behold, > the Internet. > > *Artificial General Intelligence List <https://agi.topicbox.com/latest>* > / 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/Te623675bfcbd1b22-M366367b3ef2f751be8948801> > ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/Te623675bfcbd1b22-M3fbfa33a028411b80a227046 Delivery options: https://agi.topicbox.com/groups/agi/subscription
