Zhu Xiaohu

• Zhu Xiaohu 27 Nov 2023 11:30 UTC

  LW: 1 AF: 1

  0

  AF

  on: Towards un­der­stand­ing-based safety evaluations

  I have a pretty fundamental concern with these sorts of techniques as a mechanism for eventually assessing alignment

  that would lead to safety or alignment goodharting problem.

• Zhu Xiaohu 28 Apr 2023 7:46 UTC

  8 points

  0

  in reply to: Lao Mein’s comment on: My Assess­ment of the Chi­nese AI Safety Community

  Hi. Thanks for mentioning us.

  Unlike main labs or companies in China, we are doing fundamental research work on the ontological crisis problem with model theory from mathematical logic trying to set a new base for analyzing and preventing the crisis.

  Due to our lacking of funding and restricted intellectual resources, the process is slower, but we will share our work when ready.

• Zhu Xiaohu 11 Jan 2022 9:53 UTC

  1 point

  in reply to: abramdemski’s comment on: De­ci­sion the­ory and zero-sum game the­ory, NP and PSPACE

  Mention a recent interesting work here: On the Complexity of Computing Markov Perfect Equilibrium in General-Sum Stochastic Games gave a related analysis on the comuting of Markov PE for RL agents.