Mesa-Optimization

Ruby (+1674/-62)
Ruby (+321/-1838)
riceissa (+6/-87) /* External links */
riceissa (+231) /* External links */
riceissa (+60)
riceissa (+1657) Created page with "'''Optimization daemons''' are optimizers that result from heavy optimization pressure on a different system. For example, natural selection is an optimization process (that o..."

Mesa-Optimization is the situation whichthat occurs when a learned model (such as a neural network) is itself an optimizer. A base optimizer optimizes and creates a mesa-optimizer. Previously work under this concept was called Inner Optimizer or Optimization Daemons. Arguably,

Examples

Natural selection is an optimization process (that optimizes for reproductive fitness) that produced humans (who are capable of pursuing goals that no longer correlate reliably with reproductive fitness). In this case, humans are optimization daemons of natural selection. In the context of AI alignment, the concern is that an artificial general intelligence exerting optimization pressure may produce mesa-optimizers that break alignment.1

History

Previously work under this concept was called Inner Optimizer or Optimization Daemons.

Wei Dai brings up a mesa-optimizersimilar idea in an SL4 thread.2

The optimization daemons article on Arbital was published probably in 2016.3

Jessica Taylor wrote two posts about daemons while at MIRI:

See also

References

  1. "Optimization daemons". Arbital.
  2. Wei Dai. '"friendly" humans?' December 31, 2003.

External links

Some posts that arose fromreference optimization daemons:

optimizing worst-case performance is one of the most likely reasons that I think prosaic AI alignment might turn out to be impossible (if combined with an unlucky empirical situation)." (the phrase "unlucky empirical situation" links to the optimization daemons page on Arbital)

Related ideas

Mesa-Optimization is the situation which occurs when a learned model (such as a neural network) is itself an optimizer. A base optimizer optimizes and creates a mesa-optimizer. Previously work under this concept was called Inner Optimizer or Optimization daemons are optimizers that result from heavy optimization pressure on a different system. For example, natural selection is an optimization process (that optimizes for reproductive fitness) that produced humans (who are capable of pursuing goals that no longer correlate reliably with reproductive fitness). In this case,Daemons. Arguably, humans are optimization daemonsa mesa-optimizer that arose from the base-optimizer of natural selection. In the context of AI alignment, the concern is that an artificial general intelligence exerting optimization pressure may produce daemons that break alignment.1evolution.

History

Wei Dai brings up a similar idea in an SL4 thread.2

The optimization daemons article on Arbital was published probably in 2016.3

Jessica Taylor wrote two posts about daemons while at MIRI:

See also

References

External links

Some posts that reference optimization daemons:

  • "Cause prioritization for downside-focused value systems": "Alternatively, perhaps goal preservation becomes more difficult the more capable AI systems become, in which case the future might be controlled by unstable goal functions taking turns over the steering wheel"
  • "Techniques for optimizing worst-case performance": "The difficulty of optimizing worst-case performance is one of the most likely reasons that I think prosaic AI alignment might turn out to be impossible (if combined with an unlucky empirical situation)." (the phrase "unlucky empirical situation" links to the optimization daemons page on Arbital)
  • "Prize for probable problems": "I'm happy to provisionally grant that optimization daemons would be catastrophic if you couldn’t train robust models."

Related ideas:


  1. Wei Dai. '"friendly" humans?' December 31, 2003.
  2. "Optimization daemons". Arbital.

External links

Some posts that reference optimization daemons:

Related ideas:


  1. Wei Dai. '"friendly" humans?' December 31, 2003.
  2. "Optimization daemons". Arbital.

External links

Some posts that reference optimization daemons:

Related ideas:


  1. Wei Dai. '"friendly" humans?' December 31, 2003.
  2. "Optimization daemons". Arbital.

See also

External links

Some posts that reference optimization daemons:

  • "Cause prioritization for downside-focused value systems": "Alternatively, perhaps goal preservation becomes more difficult the more capable AI systems become, in which case the future might be controlled by unstable goal functions taking turns over the steering wheel"
  • "Techniques for optimizing worst-case performance": "The difficulty of optimizing worst-case performance is one of the most likely reasons that I think prosaic AI alignment might turn out to be impossible (if combined with an unlucky empirical situation)." (the phrase "unlucky empirical situation" links to the optimization daemons page on Arbital)

Related ideas:


  1. Wei Dai. '"friendly" humans?' December 31, 2003.
  2. "Optimization daemons". Arbital.