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

Re:DeepMind says reinforcement learning is ‘enough’ to reach... by webmaven on Thursday June 10, 2021 @02:36AM (#27457553)

> RL + piggybacking on human culture might be enough, or evolution + RL for biological agents.

Yes, but over what timeframe? Will there be any diminishing returns plateaus along the way?

[ Parent | on HN ]

Re:DeepMind says reinforcement learning is ‘enough’ to reach... by visarga on Friday June 11, 2021 @02:28AM (#27469984)

We still have unknown unknowns but we also know a lot more about how neural nets deal with various tasks and dataset preparations. We know what kind of applications are good enough and where they still fail, which is much more than a decade ago.

If you look at sci-fi movies with robots, they usually speak in a metallic voice but have good situational and language understanding. In reality it was the other way around, it's much easier to do artificial voices than understand the topic. That kind of naive understanding seems silly now, and this is how we gradually advance.

GPT-3 taught us that good sounding text is not that hard to generate if you have ample training data, but modeling the larger context is still hard. These kind of fine distinctions are what I call progress.

[ Parent | on HN ]

Re:DeepMind says reinforcement learning is ‘enough’ to reach... by webmaven on Friday June 11, 2021 @09:36PM (#27480582)

> If you look at sci-fi movies with robots, they usually speak in a metallic voice but have good situational and language understanding. In reality it was the other way around, it's much easier to do artificial voices than understand the topic. That kind of naive understanding seems silly now, and this is how we gradually advance.

I waffle a lot on whether that aspect of 1968's '2001: A Space Odyssey' is evidence of genius or just survivorship bias.

[ Parent | on HN ]