Garrett Baker

SERI MATS scholar

Wiki Contributions

Comments

Re: agents terminalizing instrumental values. 

I anticipate there will be a hill-of-common-computations, where the x-axis is the frequency[1] of the instrumental subgoal, and the y-axis is the extent to which the instrumental goal has been terminalized. 

This is because for goals which are very high in frequency, there will be little incentive for the computations responsible for achieving that goal to have self-preserving structures. It will not make sense for them to devote optimization power towards ensuring future states still require them, because future states are basically guaranteed to require them.[2]

An example of this for humans may be the act of balancing while standing up. If someone offered to export this kind of cognition to a machine which did it just as good as I, I wouldn't particularly mind. If someone also wanted to change physics in such a way that the only effect is that magic invisible fairies made sure everyone stayed balancing while trying to stand up, I don't think I'd mind that either[3].

  1. ^

    I'm assuming this is frequency of the goal assuming the agent isn't optimizing to get into a state that requires that goal.

  2. ^

    This argument also assumes the overseer isn't otherwise selecting for self-preserving cognition, or that self-preserving cognition is the best way of achieving the relevant goal.

  3. ^

    Except for the part where there's magic invisible fairies in the world now. That would be cool!

The main big one was that when I was making experiments, I did not have in mind a particular theory about how the network was doing a particular capability. I just messed around with matrices, and graphed a bunch of stuff, and multiplied a bunch of weights by a bunch of other weights. Occasionally, I'd get interesting looking pictures, but I had no clue what to do with those pictures, or followup questions I could ask, and I think it's because I didn't have an explicit model of what I think it should be doing, and so couldn't update my picture of the mechanisms the network was using off the data I gathered about the network's internals.

This was really really helpful! I learned a lot about how to think through experiment design, watching you do it, and I found some possible-mistakes I've been making while designing my own experiments! 

My only criticism: When copilot auto-fills in details, it would be helpful if you'd explain what it did and why its what you wanted it to do, like how you do with your own code.