7 ms·

Based on the title, I was waiting to see mention of ML/AI but it quickly went down a path I didn't expect: the potential downsides of have AI de-anonymize data

by qudat 2y ago

Based on the title, I was waiting to see mention of ML/AI but it quickly went down a path I didn't expect: the potential downsides of have AI de-anonymize data that was previously anonymized. I'm kind of surprised by this take and ultimately think this article wasn't particularly interesting. Here's my take on the cross-section between neuroscience and AI: https://bower.sh/who-will-understand-consciousness https://bower.sh/who-will-understand-consciousness Ultimately, I don't think we are going to discover consciousness in chaotic biological systems. Rather, we will find it in silicon.

robwwilliams 2y ago
Tag team since the 1950s. Reinforcement learning is a great example of the back and fort—-Sutton and Barton’s major contribute (temporal difference learning) in computer science) followed by experimental wetware confirmation of this mechanism in CNS.
- qudat 2y ago
  Oh that's interesting! I'd love to read more collaboration between the two disciplines.
__loam 2y ago
Creating a computational simulacrum of consciousness means nothing for our understanding if we don't understand biological consciousness, it just means we've created a very capable computational system. As someone with a background in both computer science and neuroscience, the approach of abandoning the brain to go play with our toys strikes me as profoundly arrogant, and more importantly, missing the point.
- qudat 2y ago
  > if we don't understand biological consciousness, it just means we've created a very capable computational system. This is the conceit of humanity. Consciousness does not rest on us. Further, we are so inescapably biased here. A system cannot study itself and make any meaningful progress. We are shackled by the limitations of this bias, bogged down by the muck and mire of disentangling ourselves from the equation. > the approach of abandoning the brain to go play with our toys In 10 years this comment will not age well. > strikes me as profoundly arrogant, and more importantly, missing the point You've missed the point. The best way to study consciousness is by removing as much of ourselves from the equation. We must first understand consciousness through an external conscious entity. Only then can we introspect and learn something through biology.
  - __loam 2y ago
    Even if you create a "conscious" system, if you don't study the brain in vivo, you have no way of relating that "consciousness" back to our own. You've build a very intelligent system, it just had nothing to do with brains. E: if you want to talk about conceit, look no further than the idea that I'm order to study the brain, we "just" need to build one first.

by qudat 2y ago

robwwilliams 2y ago

Tag team since the 1950s. Reinforcement learning is a great example of the back and fort—-Sutton and Barton’s major contribute (temporal difference learning) in computer science) followed by experimental wetware confirmation of this mechanism in CNS.

qudat 2y ago
Oh that's interesting! I'd love to read more collaboration between the two disciplines.

__loam 2y ago

Creating a computational simulacrum of consciousness means nothing for our understanding if we don't understand biological consciousness, it just means we've created a very capable computational system. As someone with a background in both computer science and neuroscience, the approach of abandoning the brain to go play with our toys strikes me as profoundly arrogant, and more importantly, missing the point.

qudat 2y ago
> if we don't understand biological consciousness, it just means we've created a very capable computational system. This is the conceit of humanity. Consciousness does not rest on us. Further, we are so inescapably biased here. A system cannot study itself and make any meaningful progress. We are shackled by the limitations of this bias, bogged down by the muck and mire of disentangling ourselves from the equation. > the approach of abandoning the brain to go play with our toys In 10 years this comment will not age well. > strikes me as profoundly arrogant, and more importantly, missing the point You've missed the point. The best way to study consciousness is by removing as much of ourselves from the equation. We must first understand consciousness through an external conscious entity. Only then can we introspect and learn something through biology.
- __loam 2y ago
  Even if you create a "conscious" system, if you don't study the brain in vivo, you have no way of relating that "consciousness" back to our own. You've build a very intelligent system, it just had nothing to do with brains. E: if you want to talk about conceit, look no further than the idea that I'm order to study the brain, we "just" need to build one first.