3 ms·

"Deep Learning results nobody can verify" What? Deep Learning is measurably out-performing previous algorithms, and making falsifiable predictions. And if yo

"Deep Learning results nobody can verify" What? Deep Learning is measurably out-performing previous algorithms, and making falsifiable predictions. And if you read HN and can't think of progress in the last five years, then you must be incredibly cynical.

krageon 4y ago
> Deep Learning is measurably out-performing previous algorithms, and making falsifiable predictions. If you'd care to quote the entire thing where I was telling you you cannot verify it unless you have enough "mana" (i.e. money) to try it, you would also see you have answered your own rebuttal with my original post. > if you read HN and can't think of progress in the last five years, then you must be incredibly cynical. Perhaps! But I don't see you refuting me in any way, so perhaps I'm just correct :)
- VikingCoder 4y ago
  > unless you have enough "mana" (i.e. money) to try it, I use deep learning all the time, every day, and the results are better than previous algorithms. I don't need to "verify" the results to know they're better. > Perhaps! But I don't see you refuting me in any way, so perhaps I'm just correct :) You're standing in the middle of a record store and complaining that there's no good new music, and then you think the problem is that I'm not enumerating examples to refute you? If you can't find things to be impressed by on HN, then again, I think you're incredibly cynical.
Bubble_Pop_22 4y ago
> And if you read HN and can't think of progress in the last five years, then you must be incredibly cynical. Reading on HN means 99% of the times aspirational stuff. Aspirational stuff doesn't count. What counts is stuff that improves quality of life in the real world, not just aspirationally sometimes in the future in order to play the musical chair game of the stock market or VC markets. Stuff like mRNA vaccines. An other example would be fracking but that largely happened 2005-2015
- VikingCoder 4y ago
  OP also claimed "the pipeline of innovation is also empty," and then you're trying to correct me that HN is showing aspirational stuff. So which is it? Is HN filled with content showcasing the pipeline of innovation or not?
  - krageon 4y ago
    I'll thank you not to put words in my mouth.
- AstralStorm 4y ago
  We need more. We need cheap and clean energy and research in that sector has happened and is happening now, likewise water purification on large scale or we'll go dry. Improvements in farming and husbandry are also not being deployed as they're marginally more expensive, even though we critically need them. The visibly good technologies exist and need to be deployed; due to lack of investment in them, cost is prohibitive.

krageon 4y ago

> Deep Learning is measurably out-performing previous algorithms, and making falsifiable predictions. If you'd care to quote the entire thing where I was telling you you cannot verify it unless you have enough "mana" (i.e. money) to try it, you would also see you have answered your own rebuttal with my original post. > if you read HN and can't think of progress in the last five years, then you must be incredibly cynical. Perhaps! But I don't see you refuting me in any way, so perhaps I'm just correct :)

VikingCoder 4y ago
> unless you have enough "mana" (i.e. money) to try it, I use deep learning all the time, every day, and the results are better than previous algorithms. I don't need to "verify" the results to know they're better. > Perhaps! But I don't see you refuting me in any way, so perhaps I'm just correct :) You're standing in the middle of a record store and complaining that there's no good new music, and then you think the problem is that I'm not enumerating examples to refute you? If you can't find things to be impressed by on HN, then again, I think you're incredibly cynical.

Bubble_Pop_22 4y ago

> And if you read HN and can't think of progress in the last five years, then you must be incredibly cynical. Reading on HN means 99% of the times aspirational stuff. Aspirational stuff doesn't count. What counts is stuff that improves quality of life in the real world, not just aspirationally sometimes in the future in order to play the musical chair game of the stock market or VC markets. Stuff like mRNA vaccines. An other example would be fracking but that largely happened 2005-2015

VikingCoder 4y ago
OP also claimed "the pipeline of innovation is also empty," and then you're trying to correct me that HN is showing aspirational stuff. So which is it? Is HN filled with content showcasing the pipeline of innovation or not?
- krageon 4y ago
  I'll thank you not to put words in my mouth.
AstralStorm 4y ago
We need more. We need cheap and clean energy and research in that sector has happened and is happening now, likewise water purification on large scale or we'll go dry. Improvements in farming and husbandry are also not being deployed as they're marginally more expensive, even though we critically need them. The visibly good technologies exist and need to be deployed; due to lack of investment in them, cost is prohibitive.