If you run out of time, you defect. If your opponent simulates you without a time limit, and you take long enough to cause [him] to run out of time, [he’ll] defect, which isn’t what you want.
I’m not sure about that—the original post says this:
the penalty for not outputting Cooperate or Defect within the time limit has been reduced.
So presumably most bots will be set up to make a choice within the time limit.
But there’s a function that lets you run a function for up to a specified number of microseconds, and tells you what it returns or whether it times out.
Right, so the practical effect of the strategy I proposed would be to deny opponents knowledge. Of course, one can envision situations where you want to be transparent.
I’m not sure about that—the original post says this:
So presumably most bots will be set up to make a choice within the time limit.
Right, so the practical effect of the strategy I proposed would be to deny opponents knowledge. Of course, one can envision situations where you want to be transparent.