3 ms·

Alas, the world is neither reliable nor consistent, so any approach which is not probabalistic is also doomed to failure.

jqpabc123 4y ago
Yes, the advantage that human logic and reasoning offers over probabilistic AI is the ability to evaluate and make nearly instant judgments in unique circumstances without prior training. Defense driving is all about making these sort of judgments --- and choosing the least bad option when there are no good ones.
- fjkdlsjflkds 4y ago
  "ability to evaluate and make nearly instant judgments in unique circumstances without prior training" It could be argued that there are already ML systems that do what you describe (i.e. make instant effective judgements/decisions about something it hasn't been exposed to during training), at least for some contexts. For more information: https://en.wikipedia.org/wiki/Zero-shot_learning https://en.wikipedia.org/wiki/Zero-shot_learning
  - jqpabc123 4y ago
    You should tell Elon Musk so he can put this into the next sentient Tesla model.
- sdenton4 4y ago
  My view is that the Gary Marcus argument is deeply wrong. Here's why. Current methods are far from perfect, but not for the reasons Marcus believes. Consider that we /should/ be able to derive hypotheses for better system behavior from a given critique, test those proposed system improvements, and then (if the hypothesis is correct) see some improvement. Marcus' crew hasn't produced any useful improvements; this is either because they are not technically competent or their hypotheses are wrong. Given that people have been hammering at these systems for ~10 years now, I doubt the failure to build better logic-based systems (or hybrid systems) is for lack of trying. Meanwhile, the non-Marcus groups have been improving systems by leaps and bounds, through exactly the kind of iterative, hypothesis-driven improvement I described above. Certainly nothing is perfect, but we can see real progress happening. (Finally; Tesla FSD is a strawman. It's Waymo you need to argue with, and they're doing great, AFAICT.)
  - jqpabc123 4y ago
    It's Waymo you need to argue with, and they're doing great, AFAICT. Let's disperse a few thousand of them at random locations across the country for everyday use and see what happens. AFAICT, no one has actually done this yet.
    - sdenton4 4y ago
      It's a bad argument... The limited, expanding footprint is allowing them to get high reliability, exactly what you're saying you want from the systems, and which Tesla is failing to provide. Their systems are still adapting to stochastic road conditions and events, even if they aren't requiring the systems to handle them everywhere all at once. Success /anywhere/ goes a long way towards proving that the fundamental task of driving in the real world is machine learnable. There's a more interesting conversation to be had here than the one you're offering, in any case. And you haven't addressed the total failure of the Marcus crew to actually produce working solutions...

jqpabc123 4y ago

Yes, the advantage that human logic and reasoning offers over probabilistic AI is the ability to evaluate and make nearly instant judgments in unique circumstances without prior training. Defense driving is all about making these sort of judgments --- and choosing the least bad option when there are no good ones.

fjkdlsjflkds 4y ago
"ability to evaluate and make nearly instant judgments in unique circumstances without prior training" It could be argued that there are already ML systems that do what you describe (i.e. make instant effective judgements/decisions about something it hasn't been exposed to during training), at least for some contexts. For more information: https://en.wikipedia.org/wiki/Zero-shot_learning https://en.wikipedia.org/wiki/Zero-shot_learning
- jqpabc123 4y ago
  You should tell Elon Musk so he can put this into the next sentient Tesla model.
sdenton4 4y ago
My view is that the Gary Marcus argument is deeply wrong. Here's why. Current methods are far from perfect, but not for the reasons Marcus believes. Consider that we /should/ be able to derive hypotheses for better system behavior from a given critique, test those proposed system improvements, and then (if the hypothesis is correct) see some improvement. Marcus' crew hasn't produced any useful improvements; this is either because they are not technically competent or their hypotheses are wrong. Given that people have been hammering at these systems for ~10 years now, I doubt the failure to build better logic-based systems (or hybrid systems) is for lack of trying. Meanwhile, the non-Marcus groups have been improving systems by leaps and bounds, through exactly the kind of iterative, hypothesis-driven improvement I described above. Certainly nothing is perfect, but we can see real progress happening. (Finally; Tesla FSD is a strawman. It's Waymo you need to argue with, and they're doing great, AFAICT.)
- jqpabc123 4y ago
  It's Waymo you need to argue with, and they're doing great, AFAICT. Let's disperse a few thousand of them at random locations across the country for everyday use and see what happens. AFAICT, no one has actually done this yet.
  - sdenton4 4y ago
    It's a bad argument... The limited, expanding footprint is allowing them to get high reliability, exactly what you're saying you want from the systems, and which Tesla is failing to provide. Their systems are still adapting to stochastic road conditions and events, even if they aren't requiring the systems to handle them everywhere all at once. Success /anywhere/ goes a long way towards proving that the fundamental task of driving in the real world is machine learnable. There's a more interesting conversation to be had here than the one you're offering, in any case. And you haven't addressed the total failure of the Marcus crew to actually produce working solutions...