DeepMind says reinforcement learning is ‘enough’ to reach general AI

Re:DeepMind says reinforcement learning is ‘enough’ to reach... by throwaway879 on Thursday June 10, 2021 @01:58AM (#27457259)

If you send a message in a bottle it will eventually land ashore somewhere, maybe in a century, who knows, and who knows whether it will be relevant by then or not, or civilization may not even exist by then, but sure it's similarly plausible to get to AGI via RL.

[ Parent | on HN ]

Re:DeepMind says reinforcement learning is ‘enough’ to reach... by visarga on Thursday June 10, 2021 @02:24AM (#27457472)

> send a message in a bottle

Just one - yes. But how about if you send millions of bottle messages?

[ Parent | on HN ]

Re:DeepMind says reinforcement learning is ‘enough’ to reach... by throwaway879 on Thursday June 10, 2021 @05:15PM (#27465966)

>>how about if you send millions of bottle messages?

assuming we can integrate all learnings from those bottles into a system that can classify any given situation and apply the learning in that domain. But to build a system that can classify any problem is where we're stuck at and RL can't get us there

[ Parent | on HN ]

Re:DeepMind says reinforcement learning is ‘enough’ to reach... by ausbah on Thursday June 10, 2021 @02:54AM (#27457669)

then you're in your way to a hit rock song!

[ Parent | on HN ]