• Posts
  • RSS
  • ◂◂RSS
  • Contact

  • Conversation with Bryce Wiedenbeck

    July 14th, 2017
    airisk, giving  [html]
    A few days ago I spoke with Bryce Wiedenbeck, a CS professor at Swarthmore teaching AI, as part of my project of assessing superintelligence risk. Bryce had relatively similar views to Michael: AGI is possible, it could be a serious problem, but we can't productively work on it now.

    Before our conversation he looked some at Concrete Problems in AI Safety (pdf) and Deep Reinforcement Learning from Human Preferences (pdf). His view on both was that they were good work from the perspective of advancing ML but very unlikely to be relevant to making AGI safer: the systems that get us to AGI will look very different from the ones we have now.

    One reason is that he saw a lot of learning from humans as being mediated by learning utility functions, but he sees utility functions as a very limited model. Economists and others use utility functions when talking about people because that's mathematically tractable, but it's a bad description of how humans actually behave. Trying to come up with utility functions that best explain human preferences or behavior probably solves some problems nicely and is helpful, but while Bryce wouldn't completely rule it out he thought it was very unlikely to get us to AGI.

    We tried to get more into why he thinks implementations for AGI will look vastly different from what we will have today, and couldn't make progress there. Bryce thinks there are deep questions about what intelligence really is that we don't understand yet, and that as we make progress on those questions we'll develop very different sorts of ML systems. If something like today's deep learning is still a part of what we eventually end up with, it's more likely to be something that solves specific problems than as a critical component.

    (This has been a common theme in my discussions with people recently: very different intuitions on the distance to AGI in terms of technical work required, and also on whether work we're doing today is likely to transfer.)

    Comment via: google plus, facebook

    Recent posts on blogs I like:

    Willingness to look stupid

    People frequently1 think that I'm very stupid. I don't find this surprising, since I don't mind if other people think I'm stupid, which means that I don't adjust my behavior to avoid seeming stupid, which results in people thinking tha…

    via Posts on Dan Luu October 21, 2021

    Experiences in raising children in shared housing

    Sometimes I see posts about people’s hope to raise children in a group housing situation, and it often seems overly optimistic to me. In particular they seem to expect that there will be more shared childcare than I think should be expected. Today I talke…

    via The whole sky October 18, 2021

    EDT with updating double counts

    I recently got confused thinking about the following case: Calculator bet: I am offered the opportunity to bet on a mathematical statement X to which I initially assign 50% probability (perhaps X = 139926 is a quadratic residue modulo 314159). I have acce…

    via The sideways view October 12, 2021

    more     (via openring)


  • Posts
  • RSS
  • ◂◂RSS
  • Contact