6 ms·

Does world really work like that?

jsemrau 6y ago
When the first PC with Basic launched in the 80s many people wanted to develop for it. When the iPhone Appstore launched, many people started to build apps in the ecosystem. While it might be it bit too early to compare RL to those advances in technology. I personally feel there is huge potential. I might be wrong though. And I am fine with that.
- bonoboTP 6y ago
  RL isn't new though, the foundational results are about 25 years old.
  - WanderPanda 6y ago
    And it feels a bit like it is stalling (at least in continuous control)
    - cbames89 6y ago
      In my opinion there's a wide open array of approaches from control that can help with this. Learning for Control is a new conference that looks at this very topic.
  - dmarchand90 6y ago
    That's still very analogous to the first PCs. By that point there had been decades of foundational computer work
  - stevofolife 6y ago
    No one said "new". You can apply what you said to PC and iPhones. Mainframes and palms existed before them.
- bitL 6y ago
  RL needs a supercomputer and its code is usually too fragile - making a trivial mistake anywhere (missing a constant multiplication, swapping the order of two consecutive lines of code etc.) would likely lead to your model never converging even if you got everything else right.
  - Ma8ee 6y ago
    Which code is not fragile in that sense? I think that is a rather strange criticism.
  - Iv 6y ago
    You can do RL on an raspberry pi. Depends what problem you are trying to solve but not all of them require video analysis and billions of parameters.
  - cbames89 6y ago
    Technical point: Value functions that are a constant multiples of each other result in the same behavior.
    - bitL 6y ago
      Making a constant multiplication mistake somewhere in the code doesn't imply the new value function would be a constant multiply of the optimal one.
  - chasely 6y ago
    The hard part of RL for the problems I've encountered in my work is that you need a simulator. Building a reliable and accurate simulator is often an immense undertaking.
    - dgb23 6y ago
      Maybe data scientists should team up (more?) with game programmers. They have a ton of experience in building very complex simulations.

When the first PC with Basic launched in the 80s many people wanted to develop for it. When the iPhone Appstore launched, many people started to build apps in the ecosystem. While it might be it bit too early to compare RL to those advances in technology. I personally feel there is huge potential. I might be wrong though. And I am fine with that.

bonoboTP 6y ago
RL isn't new though, the foundational results are about 25 years old.
- WanderPanda 6y ago
  And it feels a bit like it is stalling (at least in continuous control)
  - cbames89 6y ago
    In my opinion there's a wide open array of approaches from control that can help with this. Learning for Control is a new conference that looks at this very topic.
- dmarchand90 6y ago
  That's still very analogous to the first PCs. By that point there had been decades of foundational computer work
- stevofolife 6y ago
  No one said "new". You can apply what you said to PC and iPhones. Mainframes and palms existed before them.
bitL 6y ago
RL needs a supercomputer and its code is usually too fragile - making a trivial mistake anywhere (missing a constant multiplication, swapping the order of two consecutive lines of code etc.) would likely lead to your model never converging even if you got everything else right.
- Ma8ee 6y ago
  Which code is not fragile in that sense? I think that is a rather strange criticism.
- Iv 6y ago
  You can do RL on an raspberry pi. Depends what problem you are trying to solve but not all of them require video analysis and billions of parameters.
- cbames89 6y ago
  Technical point: Value functions that are a constant multiples of each other result in the same behavior.
  - bitL 6y ago
    Making a constant multiplication mistake somewhere in the code doesn't imply the new value function would be a constant multiply of the optimal one.
- chasely 6y ago
  The hard part of RL for the problems I've encountered in my work is that you need a simulator. Building a reliable and accurate simulator is often an immense undertaking.
  - dgb23 6y ago
    Maybe data scientists should team up (more?) with game programmers. They have a ton of experience in building very complex simulations.