• 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:

    Learning Worst Industry Practices

    If I have a bad idea and you have a bad idea and we exchange them, we now have two bad ideas. But more than that. If I have a bad idea and you have a good idea and we exchange them, we should both land on your good idea – but that requires both […]

    via Pedestrian Observations September 20, 2020

    Collections: Iron, How Did They Make It? Part I, Mining

    This week we are starting a four-part look at pre-modern iron and steel production. As with our series on farming, we are going to follow the train of iron production from the mine to a finished object, be that a tool, a piece of armor, a simple nail, a w…

    via A Collection of Unmitigated Pedantry September 18, 2020

    Learning Game

    I came up with this game. In the game one person thinks of something and then gives the other person a clue. And the other person writes a guess down on a blackboard or a piece of paper. Or really anything you have that's laying around that's av…

    via Lily Wise's Blog Posts September 17, 2020

    more     (via openring)


  • Posts
  • RSS
  • ◂◂RSS
  • Contact