In recent years, under the direction of Luke Muehlhauser, with researchers such as Paul Christiano and the other younger guns, they may have got better, but I’m still waiting to see any technical result of theirs being published in a peer reviewed journal or conference.
We’ve released a new paper recently accepted to the MIPC workshop at AAAI-14: “Program Equilibrium in the Prisoner’s Dilemma via Löb’s Theorem” by LaVictoire et al.
We’ve released a new working paper by Benja Fallenstein and Nate Soares, “Problems of self-reference in self-improving space-time embedded intelligence.” [...]
Update 05/14/14: This paper has been accepted to AGI-14.
http://intelligence.org/2014/05/17/new-paper-program-equilibrium-prisoners-dilemma-via-lobs-theorem/ :
http://intelligence.org/2014/05/06/new-paper-problems-of-self-reference-in-self-improving-space-time-embedded-intelligence/ :
Didn’t know about that. Thanks for the update.