4 ms·

Improving performance for a metric that we actually know how to quantify is fundamentally different than calling interesting but potentially one-off stabs as in

Improving performance for a metric that we actually know how to quantify is fundamentally different than calling interesting but potentially one-off stabs as incremental progress. Conflating them only demonstrates how far we have to go.

unishark 6y ago
We do know how to quantify intelligence (unless you're one of those who includes consciousness in the definition of AGI, which seems almost religious to me). I expect it will quickly switch to an argument of what deficiencies AGI candidates have and how smart they really are, much like ML models that supposedly beat humans at narrow tasks today, rather than possible versus impossible talk. As for impossible talk, we have biological examples all around us of what needs to be built. We just need to imitate. Much like computer vision, algorithms sucked at it until they didn't (and all it took was someone scaling up an old design idea plus a lot of data). On the scale of gigantic ambitious goals it's pretty special in that regard. Curing cancer or death or mars colonies may indeed be impossible, by contrast. I will agree that I trust no one's ability to predict anything. They are all just making almost-entirely-uneducated guess using a few variables out of some vast number of unknowns.
- visarga 6y ago
  > unless you're one of those who includes consciousness in the definition of AGI Why stick with this concept - consciousness - which is not well defined, instead of using a much more practical concept: embodiment. Embodiment is the missing link towards human level AI. Agents embodied in a world, like AlphaGO, already surpass humans (on the Go board or Dota 2), we just need to take that ability to real world. The source of meaning is in the game, not the brain. What we need is a better simulator of the world, or a neural technique for imagination in RL, which is under works [1]. 1. https://arxiv.org/pdf/2005.05960.pdf https://arxiv.org/pdf/2005.05960.pdf
- loosetypes 6y ago
  Thanks for the reply. > We do know how to quantify intelligence (..). But how exactly do we do that?

unishark 6y ago

We do know how to quantify intelligence (unless you're one of those who includes consciousness in the definition of AGI, which seems almost religious to me). I expect it will quickly switch to an argument of what deficiencies AGI candidates have and how smart they really are, much like ML models that supposedly beat humans at narrow tasks today, rather than possible versus impossible talk. As for impossible talk, we have biological examples all around us of what needs to be built. We just need to imitate. Much like computer vision, algorithms sucked at it until they didn't (and all it took was someone scaling up an old design idea plus a lot of data). On the scale of gigantic ambitious goals it's pretty special in that regard. Curing cancer or death or mars colonies may indeed be impossible, by contrast. I will agree that I trust no one's ability to predict anything. They are all just making almost-entirely-uneducated guess using a few variables out of some vast number of unknowns.

visarga 6y ago
> unless you're one of those who includes consciousness in the definition of AGI Why stick with this concept - consciousness - which is not well defined, instead of using a much more practical concept: embodiment. Embodiment is the missing link towards human level AI. Agents embodied in a world, like AlphaGO, already surpass humans (on the Go board or Dota 2), we just need to take that ability to real world. The source of meaning is in the game, not the brain. What we need is a better simulator of the world, or a neural technique for imagination in RL, which is under works [1]. 1. https://arxiv.org/pdf/2005.05960.pdf https://arxiv.org/pdf/2005.05960.pdf
loosetypes 6y ago
Thanks for the reply. > We do know how to quantify intelligence (..). But how exactly do we do that?