There are (and should be) many different ways to assess generalization, but you might start with @LakeBrenden’s updates to my late 90’s work. And look at @dileeplearning’s work on how DQN fails in tiny changes to Atari games.
-
Internal products are still products, $100M is noteworthy given the basic research focus. The notion of using revenue/costs to assess scientific merit is rather unique to you. What lab is doing better by this metric? Has it historically been a precursor to larger innovation?
Just to clarify from my previous posts: I wasn't saying that DeepMind doesn't hype their work; they absolutely do. I was just questioning your criticism that they don't have a successful deep RL *commercial product*.
Ps why sell outside? For the validation that the technology is real and generalizeable. Business world generally doesn’t see internal sales as full validation, with good reason: if your rich uncle buys your lemonade, it doesn’t mean you’ve perfected the formula.
-
-
Again, what lab is succeeding by this metric?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
So owing your uncle money is real, but the uncle paying you is not? I agree internal finances aren't the same as external, but the article is about the former.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.