4 ms·

The Boston Dynamics PR machine is in full swing. A recent acquisition, the 60 minutes interview last weekend, this recent press release... Seems like they need

by anonu 6y ago

The Boston Dynamics PR machine is in full swing. A recent acquisition, the 60 minutes interview last weekend, this recent press release... Seems like they need to figure out how to make some revenues.

amelius 6y ago
The problem is they based all their tech on old-fashioned control theory, which will have its ass kicked by deep-learning-based tech when the first serious competitor comes along.
- j-pb 6y ago
  Doubtful. You need provable security and systems characteristics if you wanted to get approval for a household robot in for example the E.U. Higher cognition and task solving? Sure, we can do that with blackbox methods, but everything where skulls could get crushed, you want, and by law need, the boring stuff.
  - amelius 6y ago
    Meanwhile, Tesla's Autopilot is using deep learning to detect roads, cars, objects and people.
    - nullifidian 6y ago
      And it has to be supervised by the driver.
    - jandrese 6y ago
      And still doing a poor job of it.
      - amelius 6y ago
        You can reverse the reasoning: if you can't safely manuoever a car through a street with people, then how do you expect to do the same with a legged robot?
        Aeronwen 6y ago
        It's slow enough to avoid by walking around it, and you can make it small and light enough to not harm someone it might bump into.
        amelius 6y ago
        Yes, this applies both to control-theory and to deep learning. So I'd say choose the most flexible approach, which is deep learning.
    - j-pb 6y ago
      I'm not saying that machine learning isn't capable of doing these things (most of the time). I'm saying there's no way in hell you'll get a non compliant blackbox system past European safety laws.
      - amelius 6y ago
        But robots need to do more than make accurate movements. They need to detect where they can safely move. For this last part, you can't realistically make provably correct control algorithms (unless you disallow interaction with humans). Hence, you might as well use deep learning for the entire system. Also, you can limit the damage the system can make in some other way (e.g. limit the acceleration of the robot or make it lightweight, etc.)

by anonu 6y ago

amelius 6y ago

The problem is they based all their tech on old-fashioned control theory, which will have its ass kicked by deep-learning-based tech when the first serious competitor comes along.

j-pb 6y ago
Doubtful. You need provable security and systems characteristics if you wanted to get approval for a household robot in for example the E.U. Higher cognition and task solving? Sure, we can do that with blackbox methods, but everything where skulls could get crushed, you want, and by law need, the boring stuff.
- amelius 6y ago
  Meanwhile, Tesla's Autopilot is using deep learning to detect roads, cars, objects and people.
  - nullifidian 6y ago
    And it has to be supervised by the driver.
  - jandrese 6y ago
    And still doing a poor job of it.
    - amelius 6y ago
      You can reverse the reasoning: if you can't safely manuoever a car through a street with people, then how do you expect to do the same with a legged robot?
      - Aeronwen 6y ago
        It's slow enough to avoid by walking around it, and you can make it small and light enough to not harm someone it might bump into.
        amelius 6y ago
        Yes, this applies both to control-theory and to deep learning. So I'd say choose the most flexible approach, which is deep learning.
  - j-pb 6y ago
    I'm not saying that machine learning isn't capable of doing these things (most of the time). I'm saying there's no way in hell you'll get a non compliant blackbox system past European safety laws.
    - amelius 6y ago
      But robots need to do more than make accurate movements. They need to detect where they can safely move. For this last part, you can't realistically make provably correct control algorithms (unless you disallow interaction with humans). Hence, you might as well use deep learning for the entire system. Also, you can limit the damage the system can make in some other way (e.g. limit the acceleration of the robot or make it lightweight, etc.)